📚 Documentação inicial do ALETHEIA

- MANUAL-PRODUTO.md: Manual do usuário final
- MANUAL-VENDAS.md: Estratégia comercial e vendas
- MANUAL-TECNICO.md: Infraestrutura e deploy
- README.md: Visão geral do projeto

backend/venv/lib/python3.12/site-packages/openai/resources/__init__.py

@@ -0,0 +1,173 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from .beta import (
+    Beta,
+    AsyncBeta,
+    BetaWithRawResponse,
+    AsyncBetaWithRawResponse,
+    BetaWithStreamingResponse,
+    AsyncBetaWithStreamingResponse,
+)
+from .chat import (
+    Chat,
+    AsyncChat,
+    ChatWithRawResponse,
+    AsyncChatWithRawResponse,
+    ChatWithStreamingResponse,
+    AsyncChatWithStreamingResponse,
+)
+from .audio import (
+    Audio,
+    AsyncAudio,
+    AudioWithRawResponse,
+    AsyncAudioWithRawResponse,
+    AudioWithStreamingResponse,
+    AsyncAudioWithStreamingResponse,
+)
+from .files import (
+    Files,
+    AsyncFiles,
+    FilesWithRawResponse,
+    AsyncFilesWithRawResponse,
+    FilesWithStreamingResponse,
+    AsyncFilesWithStreamingResponse,
+)
+from .images import (
+    Images,
+    AsyncImages,
+    ImagesWithRawResponse,
+    AsyncImagesWithRawResponse,
+    ImagesWithStreamingResponse,
+    AsyncImagesWithStreamingResponse,
+)
+from .models import (
+    Models,
+    AsyncModels,
+    ModelsWithRawResponse,
+    AsyncModelsWithRawResponse,
+    ModelsWithStreamingResponse,
+    AsyncModelsWithStreamingResponse,
+)
+from .batches import (
+    Batches,
+    AsyncBatches,
+    BatchesWithRawResponse,
+    AsyncBatchesWithRawResponse,
+    BatchesWithStreamingResponse,
+    AsyncBatchesWithStreamingResponse,
+)
+from .uploads import (
+    Uploads,
+    AsyncUploads,
+    UploadsWithRawResponse,
+    AsyncUploadsWithRawResponse,
+    UploadsWithStreamingResponse,
+    AsyncUploadsWithStreamingResponse,
+)
+from .embeddings import (
+    Embeddings,
+    AsyncEmbeddings,
+    EmbeddingsWithRawResponse,
+    AsyncEmbeddingsWithRawResponse,
+    EmbeddingsWithStreamingResponse,
+    AsyncEmbeddingsWithStreamingResponse,
+)
+from .completions import (
+    Completions,
+    AsyncCompletions,
+    CompletionsWithRawResponse,
+    AsyncCompletionsWithRawResponse,
+    CompletionsWithStreamingResponse,
+    AsyncCompletionsWithStreamingResponse,
+)
+from .fine_tuning import (
+    FineTuning,
+    AsyncFineTuning,
+    FineTuningWithRawResponse,
+    AsyncFineTuningWithRawResponse,
+    FineTuningWithStreamingResponse,
+    AsyncFineTuningWithStreamingResponse,
+)
+from .moderations import (
+    Moderations,
+    AsyncModerations,
+    ModerationsWithRawResponse,
+    AsyncModerationsWithRawResponse,
+    ModerationsWithStreamingResponse,
+    AsyncModerationsWithStreamingResponse,
+)
+
+__all__ = [
+    "Completions",
+    "AsyncCompletions",
+    "CompletionsWithRawResponse",
+    "AsyncCompletionsWithRawResponse",
+    "CompletionsWithStreamingResponse",
+    "AsyncCompletionsWithStreamingResponse",
+    "Chat",
+    "AsyncChat",
+    "ChatWithRawResponse",
+    "AsyncChatWithRawResponse",
+    "ChatWithStreamingResponse",
+    "AsyncChatWithStreamingResponse",
+    "Embeddings",
+    "AsyncEmbeddings",
+    "EmbeddingsWithRawResponse",
+    "AsyncEmbeddingsWithRawResponse",
+    "EmbeddingsWithStreamingResponse",
+    "AsyncEmbeddingsWithStreamingResponse",
+    "Files",
+    "AsyncFiles",
+    "FilesWithRawResponse",
+    "AsyncFilesWithRawResponse",
+    "FilesWithStreamingResponse",
+    "AsyncFilesWithStreamingResponse",
+    "Images",
+    "AsyncImages",
+    "ImagesWithRawResponse",
+    "AsyncImagesWithRawResponse",
+    "ImagesWithStreamingResponse",
+    "AsyncImagesWithStreamingResponse",
+    "Audio",
+    "AsyncAudio",
+    "AudioWithRawResponse",
+    "AsyncAudioWithRawResponse",
+    "AudioWithStreamingResponse",
+    "AsyncAudioWithStreamingResponse",
+    "Moderations",
+    "AsyncModerations",
+    "ModerationsWithRawResponse",
+    "AsyncModerationsWithRawResponse",
+    "ModerationsWithStreamingResponse",
+    "AsyncModerationsWithStreamingResponse",
+    "Models",
+    "AsyncModels",
+    "ModelsWithRawResponse",
+    "AsyncModelsWithRawResponse",
+    "ModelsWithStreamingResponse",
+    "AsyncModelsWithStreamingResponse",
+    "FineTuning",
+    "AsyncFineTuning",
+    "FineTuningWithRawResponse",
+    "AsyncFineTuningWithRawResponse",
+    "FineTuningWithStreamingResponse",
+    "AsyncFineTuningWithStreamingResponse",
+    "Beta",
+    "AsyncBeta",
+    "BetaWithRawResponse",
+    "AsyncBetaWithRawResponse",
+    "BetaWithStreamingResponse",
+    "AsyncBetaWithStreamingResponse",
+    "Batches",
+    "AsyncBatches",
+    "BatchesWithRawResponse",
+    "AsyncBatchesWithRawResponse",
+    "BatchesWithStreamingResponse",
+    "AsyncBatchesWithStreamingResponse",
+    "Uploads",
+    "AsyncUploads",
+    "UploadsWithRawResponse",
+    "AsyncUploadsWithRawResponse",
+    "UploadsWithStreamingResponse",
+    "AsyncUploadsWithStreamingResponse",
+]

backend/venv/lib/python3.12/site-packages/openai/resources/audio/__init__.py

@@ -0,0 +1,61 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from .audio import (
+    Audio,
+    AsyncAudio,
+    AudioWithRawResponse,
+    AsyncAudioWithRawResponse,
+    AudioWithStreamingResponse,
+    AsyncAudioWithStreamingResponse,
+)
+from .speech import (
+    Speech,
+    AsyncSpeech,
+    SpeechWithRawResponse,
+    AsyncSpeechWithRawResponse,
+    SpeechWithStreamingResponse,
+    AsyncSpeechWithStreamingResponse,
+)
+from .translations import (
+    Translations,
+    AsyncTranslations,
+    TranslationsWithRawResponse,
+    AsyncTranslationsWithRawResponse,
+    TranslationsWithStreamingResponse,
+    AsyncTranslationsWithStreamingResponse,
+)
+from .transcriptions import (
+    Transcriptions,
+    AsyncTranscriptions,
+    TranscriptionsWithRawResponse,
+    AsyncTranscriptionsWithRawResponse,
+    TranscriptionsWithStreamingResponse,
+    AsyncTranscriptionsWithStreamingResponse,
+)
+
+__all__ = [
+    "Transcriptions",
+    "AsyncTranscriptions",
+    "TranscriptionsWithRawResponse",
+    "AsyncTranscriptionsWithRawResponse",
+    "TranscriptionsWithStreamingResponse",
+    "AsyncTranscriptionsWithStreamingResponse",
+    "Translations",
+    "AsyncTranslations",
+    "TranslationsWithRawResponse",
+    "AsyncTranslationsWithRawResponse",
+    "TranslationsWithStreamingResponse",
+    "AsyncTranslationsWithStreamingResponse",
+    "Speech",
+    "AsyncSpeech",
+    "SpeechWithRawResponse",
+    "AsyncSpeechWithRawResponse",
+    "SpeechWithStreamingResponse",
+    "AsyncSpeechWithStreamingResponse",
+    "Audio",
+    "AsyncAudio",
+    "AudioWithRawResponse",
+    "AsyncAudioWithRawResponse",
+    "AudioWithStreamingResponse",
+    "AsyncAudioWithStreamingResponse",
+]

backend/venv/lib/python3.12/site-packages/openai/resources/audio/audio.py

@@ -0,0 +1,166 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from .speech import (
+    Speech,
+    AsyncSpeech,
+    SpeechWithRawResponse,
+    AsyncSpeechWithRawResponse,
+    SpeechWithStreamingResponse,
+    AsyncSpeechWithStreamingResponse,
+)
+from ..._compat import cached_property
+from ..._resource import SyncAPIResource, AsyncAPIResource
+from .translations import (
+    Translations,
+    AsyncTranslations,
+    TranslationsWithRawResponse,
+    AsyncTranslationsWithRawResponse,
+    TranslationsWithStreamingResponse,
+    AsyncTranslationsWithStreamingResponse,
+)
+from .transcriptions import (
+    Transcriptions,
+    AsyncTranscriptions,
+    TranscriptionsWithRawResponse,
+    AsyncTranscriptionsWithRawResponse,
+    TranscriptionsWithStreamingResponse,
+    AsyncTranscriptionsWithStreamingResponse,
+)
+
+__all__ = ["Audio", "AsyncAudio"]
+
+
+class Audio(SyncAPIResource):
+    @cached_property
+    def transcriptions(self) -> Transcriptions:
+        return Transcriptions(self._client)
+
+    @cached_property
+    def translations(self) -> Translations:
+        return Translations(self._client)
+
+    @cached_property
+    def speech(self) -> Speech:
+        return Speech(self._client)
+
+    @cached_property
+    def with_raw_response(self) -> AudioWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AudioWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AudioWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AudioWithStreamingResponse(self)
+
+
+class AsyncAudio(AsyncAPIResource):
+    @cached_property
+    def transcriptions(self) -> AsyncTranscriptions:
+        return AsyncTranscriptions(self._client)
+
+    @cached_property
+    def translations(self) -> AsyncTranslations:
+        return AsyncTranslations(self._client)
+
+    @cached_property
+    def speech(self) -> AsyncSpeech:
+        return AsyncSpeech(self._client)
+
+    @cached_property
+    def with_raw_response(self) -> AsyncAudioWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncAudioWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncAudioWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncAudioWithStreamingResponse(self)
+
+
+class AudioWithRawResponse:
+    def __init__(self, audio: Audio) -> None:
+        self._audio = audio
+
+    @cached_property
+    def transcriptions(self) -> TranscriptionsWithRawResponse:
+        return TranscriptionsWithRawResponse(self._audio.transcriptions)
+
+    @cached_property
+    def translations(self) -> TranslationsWithRawResponse:
+        return TranslationsWithRawResponse(self._audio.translations)
+
+    @cached_property
+    def speech(self) -> SpeechWithRawResponse:
+        return SpeechWithRawResponse(self._audio.speech)
+
+
+class AsyncAudioWithRawResponse:
+    def __init__(self, audio: AsyncAudio) -> None:
+        self._audio = audio
+
+    @cached_property
+    def transcriptions(self) -> AsyncTranscriptionsWithRawResponse:
+        return AsyncTranscriptionsWithRawResponse(self._audio.transcriptions)
+
+    @cached_property
+    def translations(self) -> AsyncTranslationsWithRawResponse:
+        return AsyncTranslationsWithRawResponse(self._audio.translations)
+
+    @cached_property
+    def speech(self) -> AsyncSpeechWithRawResponse:
+        return AsyncSpeechWithRawResponse(self._audio.speech)
+
+
+class AudioWithStreamingResponse:
+    def __init__(self, audio: Audio) -> None:
+        self._audio = audio
+
+    @cached_property
+    def transcriptions(self) -> TranscriptionsWithStreamingResponse:
+        return TranscriptionsWithStreamingResponse(self._audio.transcriptions)
+
+    @cached_property
+    def translations(self) -> TranslationsWithStreamingResponse:
+        return TranslationsWithStreamingResponse(self._audio.translations)
+
+    @cached_property
+    def speech(self) -> SpeechWithStreamingResponse:
+        return SpeechWithStreamingResponse(self._audio.speech)
+
+
+class AsyncAudioWithStreamingResponse:
+    def __init__(self, audio: AsyncAudio) -> None:
+        self._audio = audio
+
+    @cached_property
+    def transcriptions(self) -> AsyncTranscriptionsWithStreamingResponse:
+        return AsyncTranscriptionsWithStreamingResponse(self._audio.transcriptions)
+
+    @cached_property
+    def translations(self) -> AsyncTranslationsWithStreamingResponse:
+        return AsyncTranslationsWithStreamingResponse(self._audio.translations)
+
+    @cached_property
+    def speech(self) -> AsyncSpeechWithStreamingResponse:
+        return AsyncSpeechWithStreamingResponse(self._audio.speech)

backend/venv/lib/python3.12/site-packages/openai/resources/audio/speech.py

@@ -0,0 +1,234 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Union
+from typing_extensions import Literal
+
+import httpx
+
+from ... import _legacy_response
+from ..._types import NOT_GIVEN, Body, Query, Headers, NotGiven
+from ..._utils import (
+    maybe_transform,
+    async_maybe_transform,
+)
+from ..._compat import cached_property
+from ..._resource import SyncAPIResource, AsyncAPIResource
+from ..._response import (
+    StreamedBinaryAPIResponse,
+    AsyncStreamedBinaryAPIResponse,
+    to_custom_streamed_response_wrapper,
+    async_to_custom_streamed_response_wrapper,
+)
+from ...types.audio import speech_create_params
+from ..._base_client import make_request_options
+from ...types.audio.speech_model import SpeechModel
+
+__all__ = ["Speech", "AsyncSpeech"]
+
+
+class Speech(SyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> SpeechWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return SpeechWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> SpeechWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return SpeechWithStreamingResponse(self)
+
+    def create(
+        self,
+        *,
+        input: str,
+        model: Union[str, SpeechModel],
+        voice: Literal["alloy", "echo", "fable", "onyx", "nova", "shimmer"],
+        response_format: Literal["mp3", "opus", "aac", "flac", "wav", "pcm"] | NotGiven = NOT_GIVEN,
+        speed: float | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> _legacy_response.HttpxBinaryResponseContent:
+        """
+        Generates audio from the input text.
+
+        Args:
+          input: The text to generate audio for. The maximum length is 4096 characters.
+
+          model:
+              One of the available [TTS models](https://platform.openai.com/docs/models/tts):
+              `tts-1` or `tts-1-hd`
+
+          voice: The voice to use when generating the audio. Supported voices are `alloy`,
+              `echo`, `fable`, `onyx`, `nova`, and `shimmer`. Previews of the voices are
+              available in the
+              [Text to speech guide](https://platform.openai.com/docs/guides/text-to-speech/voice-options).
+
+          response_format: The format to audio in. Supported formats are `mp3`, `opus`, `aac`, `flac`,
+              `wav`, and `pcm`.
+
+          speed: The speed of the generated audio. Select a value from `0.25` to `4.0`. `1.0` is
+              the default.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        extra_headers = {"Accept": "application/octet-stream", **(extra_headers or {})}
+        return self._post(
+            "/audio/speech",
+            body=maybe_transform(
+                {
+                    "input": input,
+                    "model": model,
+                    "voice": voice,
+                    "response_format": response_format,
+                    "speed": speed,
+                },
+                speech_create_params.SpeechCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=_legacy_response.HttpxBinaryResponseContent,
+        )
+
+
+class AsyncSpeech(AsyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> AsyncSpeechWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncSpeechWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncSpeechWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncSpeechWithStreamingResponse(self)
+
+    async def create(
+        self,
+        *,
+        input: str,
+        model: Union[str, SpeechModel],
+        voice: Literal["alloy", "echo", "fable", "onyx", "nova", "shimmer"],
+        response_format: Literal["mp3", "opus", "aac", "flac", "wav", "pcm"] | NotGiven = NOT_GIVEN,
+        speed: float | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> _legacy_response.HttpxBinaryResponseContent:
+        """
+        Generates audio from the input text.
+
+        Args:
+          input: The text to generate audio for. The maximum length is 4096 characters.
+
+          model:
+              One of the available [TTS models](https://platform.openai.com/docs/models/tts):
+              `tts-1` or `tts-1-hd`
+
+          voice: The voice to use when generating the audio. Supported voices are `alloy`,
+              `echo`, `fable`, `onyx`, `nova`, and `shimmer`. Previews of the voices are
+              available in the
+              [Text to speech guide](https://platform.openai.com/docs/guides/text-to-speech/voice-options).
+
+          response_format: The format to audio in. Supported formats are `mp3`, `opus`, `aac`, `flac`,
+              `wav`, and `pcm`.
+
+          speed: The speed of the generated audio. Select a value from `0.25` to `4.0`. `1.0` is
+              the default.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        extra_headers = {"Accept": "application/octet-stream", **(extra_headers or {})}
+        return await self._post(
+            "/audio/speech",
+            body=await async_maybe_transform(
+                {
+                    "input": input,
+                    "model": model,
+                    "voice": voice,
+                    "response_format": response_format,
+                    "speed": speed,
+                },
+                speech_create_params.SpeechCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=_legacy_response.HttpxBinaryResponseContent,
+        )
+
+
+class SpeechWithRawResponse:
+    def __init__(self, speech: Speech) -> None:
+        self._speech = speech
+
+        self.create = _legacy_response.to_raw_response_wrapper(
+            speech.create,
+        )
+
+
+class AsyncSpeechWithRawResponse:
+    def __init__(self, speech: AsyncSpeech) -> None:
+        self._speech = speech
+
+        self.create = _legacy_response.async_to_raw_response_wrapper(
+            speech.create,
+        )
+
+
+class SpeechWithStreamingResponse:
+    def __init__(self, speech: Speech) -> None:
+        self._speech = speech
+
+        self.create = to_custom_streamed_response_wrapper(
+            speech.create,
+            StreamedBinaryAPIResponse,
+        )
+
+
+class AsyncSpeechWithStreamingResponse:
+    def __init__(self, speech: AsyncSpeech) -> None:
+        self._speech = speech
+
+        self.create = async_to_custom_streamed_response_wrapper(
+            speech.create,
+            AsyncStreamedBinaryAPIResponse,
+        )

backend/venv/lib/python3.12/site-packages/openai/resources/audio/transcriptions.py

@@ -0,0 +1,415 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+import logging
+from typing import TYPE_CHECKING, List, Union, Mapping, cast
+from typing_extensions import Literal, overload, assert_never
+
+import httpx
+
+from ... import _legacy_response
+from ...types import AudioResponseFormat
+from ..._types import NOT_GIVEN, Body, Query, Headers, NotGiven, FileTypes
+from ..._utils import (
+    extract_files,
+    maybe_transform,
+    deepcopy_minimal,
+    async_maybe_transform,
+)
+from ..._compat import cached_property
+from ..._resource import SyncAPIResource, AsyncAPIResource
+from ..._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
+from ...types.audio import transcription_create_params
+from ..._base_client import make_request_options
+from ...types.audio_model import AudioModel
+from ...types.audio.transcription import Transcription
+from ...types.audio_response_format import AudioResponseFormat
+from ...types.audio.transcription_verbose import TranscriptionVerbose
+
+__all__ = ["Transcriptions", "AsyncTranscriptions"]
+
+log: logging.Logger = logging.getLogger("openai.audio.transcriptions")
+
+
+class Transcriptions(SyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> TranscriptionsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return TranscriptionsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> TranscriptionsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return TranscriptionsWithStreamingResponse(self)
+
+    @overload
+    def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        response_format: Union[Literal["json"], NotGiven] = NOT_GIVEN,
+        language: str | NotGiven = NOT_GIVEN,
+        prompt: str | NotGiven = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        timestamp_granularities: List[Literal["word", "segment"]] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Transcription: ...
+
+    @overload
+    def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        response_format: Literal["verbose_json"],
+        language: str | NotGiven = NOT_GIVEN,
+        prompt: str | NotGiven = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        timestamp_granularities: List[Literal["word", "segment"]] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> TranscriptionVerbose: ...
+
+    @overload
+    def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        response_format: Literal["text", "srt", "vtt"],
+        language: str | NotGiven = NOT_GIVEN,
+        prompt: str | NotGiven = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        timestamp_granularities: List[Literal["word", "segment"]] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> str: ...
+
+    def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        language: str | NotGiven = NOT_GIVEN,
+        prompt: str | NotGiven = NOT_GIVEN,
+        response_format: Union[AudioResponseFormat, NotGiven] = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        timestamp_granularities: List[Literal["word", "segment"]] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Transcription | TranscriptionVerbose | str:
+        """
+        Transcribes audio into the input language.
+
+        Args:
+          file:
+              The audio file object (not file name) to transcribe, in one of these formats:
+              flac, mp3, mp4, mpeg, mpga, m4a, ogg, wav, or webm.
+
+          model: ID of the model to use. Only `whisper-1` (which is powered by our open source
+              Whisper V2 model) is currently available.
+
+          language: The language of the input audio. Supplying the input language in
+              [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) format will
+              improve accuracy and latency.
+
+          prompt: An optional text to guide the model's style or continue a previous audio
+              segment. The
+              [prompt](https://platform.openai.com/docs/guides/speech-to-text/prompting)
+              should match the audio language.
+
+          response_format: The format of the output, in one of these options: `json`, `text`, `srt`,
+              `verbose_json`, or `vtt`.
+
+          temperature: The sampling temperature, between 0 and 1. Higher values like 0.8 will make the
+              output more random, while lower values like 0.2 will make it more focused and
+              deterministic. If set to 0, the model will use
+              [log probability](https://en.wikipedia.org/wiki/Log_probability) to
+              automatically increase the temperature until certain thresholds are hit.
+
+          timestamp_granularities: The timestamp granularities to populate for this transcription.
+              `response_format` must be set `verbose_json` to use timestamp granularities.
+              Either or both of these options are supported: `word`, or `segment`. Note: There
+              is no additional latency for segment timestamps, but generating word timestamps
+              incurs additional latency.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        body = deepcopy_minimal(
+            {
+                "file": file,
+                "model": model,
+                "language": language,
+                "prompt": prompt,
+                "response_format": response_format,
+                "temperature": temperature,
+                "timestamp_granularities": timestamp_granularities,
+            }
+        )
+        files = extract_files(cast(Mapping[str, object], body), paths=[["file"]])
+        # It should be noted that the actual Content-Type header that will be
+        # sent to the server will contain a `boundary` parameter, e.g.
+        # multipart/form-data; boundary=---abc--
+        extra_headers = {"Content-Type": "multipart/form-data", **(extra_headers or {})}
+        return self._post(  # type: ignore[return-value]
+            "/audio/transcriptions",
+            body=maybe_transform(body, transcription_create_params.TranscriptionCreateParams),
+            files=files,
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=_get_response_format_type(response_format),
+        )
+
+
+class AsyncTranscriptions(AsyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> AsyncTranscriptionsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncTranscriptionsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncTranscriptionsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncTranscriptionsWithStreamingResponse(self)
+
+    @overload
+    async def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        response_format: Union[Literal["json"], NotGiven] = NOT_GIVEN,
+        language: str | NotGiven = NOT_GIVEN,
+        prompt: str | NotGiven = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        timestamp_granularities: List[Literal["word", "segment"]] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Transcription: ...
+
+    @overload
+    async def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        response_format: Literal["verbose_json"],
+        language: str | NotGiven = NOT_GIVEN,
+        prompt: str | NotGiven = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        timestamp_granularities: List[Literal["word", "segment"]] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> TranscriptionVerbose: ...
+
+    @overload
+    async def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        response_format: Literal["text", "srt", "vtt"],
+        language: str | NotGiven = NOT_GIVEN,
+        prompt: str | NotGiven = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        timestamp_granularities: List[Literal["word", "segment"]] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> str: ...
+
+    async def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        language: str | NotGiven = NOT_GIVEN,
+        prompt: str | NotGiven = NOT_GIVEN,
+        response_format: Union[AudioResponseFormat, NotGiven] = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        timestamp_granularities: List[Literal["word", "segment"]] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Transcription | TranscriptionVerbose | str:
+        """
+        Transcribes audio into the input language.
+
+        Args:
+          file:
+              The audio file object (not file name) to transcribe, in one of these formats:
+              flac, mp3, mp4, mpeg, mpga, m4a, ogg, wav, or webm.
+
+          model: ID of the model to use. Only `whisper-1` (which is powered by our open source
+              Whisper V2 model) is currently available.
+
+          language: The language of the input audio. Supplying the input language in
+              [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) format will
+              improve accuracy and latency.
+
+          prompt: An optional text to guide the model's style or continue a previous audio
+              segment. The
+              [prompt](https://platform.openai.com/docs/guides/speech-to-text/prompting)
+              should match the audio language.
+
+          response_format: The format of the output, in one of these options: `json`, `text`, `srt`,
+              `verbose_json`, or `vtt`.
+
+          temperature: The sampling temperature, between 0 and 1. Higher values like 0.8 will make the
+              output more random, while lower values like 0.2 will make it more focused and
+              deterministic. If set to 0, the model will use
+              [log probability](https://en.wikipedia.org/wiki/Log_probability) to
+              automatically increase the temperature until certain thresholds are hit.
+
+          timestamp_granularities: The timestamp granularities to populate for this transcription.
+              `response_format` must be set `verbose_json` to use timestamp granularities.
+              Either or both of these options are supported: `word`, or `segment`. Note: There
+              is no additional latency for segment timestamps, but generating word timestamps
+              incurs additional latency.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        body = deepcopy_minimal(
+            {
+                "file": file,
+                "model": model,
+                "language": language,
+                "prompt": prompt,
+                "response_format": response_format,
+                "temperature": temperature,
+                "timestamp_granularities": timestamp_granularities,
+            }
+        )
+        files = extract_files(cast(Mapping[str, object], body), paths=[["file"]])
+        # It should be noted that the actual Content-Type header that will be
+        # sent to the server will contain a `boundary` parameter, e.g.
+        # multipart/form-data; boundary=---abc--
+        extra_headers = {"Content-Type": "multipart/form-data", **(extra_headers or {})}
+        return await self._post(
+            "/audio/transcriptions",
+            body=await async_maybe_transform(body, transcription_create_params.TranscriptionCreateParams),
+            files=files,
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=_get_response_format_type(response_format),
+        )
+
+
+class TranscriptionsWithRawResponse:
+    def __init__(self, transcriptions: Transcriptions) -> None:
+        self._transcriptions = transcriptions
+
+        self.create = _legacy_response.to_raw_response_wrapper(
+            transcriptions.create,
+        )
+
+
+class AsyncTranscriptionsWithRawResponse:
+    def __init__(self, transcriptions: AsyncTranscriptions) -> None:
+        self._transcriptions = transcriptions
+
+        self.create = _legacy_response.async_to_raw_response_wrapper(
+            transcriptions.create,
+        )
+
+
+class TranscriptionsWithStreamingResponse:
+    def __init__(self, transcriptions: Transcriptions) -> None:
+        self._transcriptions = transcriptions
+
+        self.create = to_streamed_response_wrapper(
+            transcriptions.create,
+        )
+
+
+class AsyncTranscriptionsWithStreamingResponse:
+    def __init__(self, transcriptions: AsyncTranscriptions) -> None:
+        self._transcriptions = transcriptions
+
+        self.create = async_to_streamed_response_wrapper(
+            transcriptions.create,
+        )
+
+
+def _get_response_format_type(
+    response_format: Literal["json", "text", "srt", "verbose_json", "vtt"] | NotGiven,
+) -> type[Transcription | TranscriptionVerbose | str]:
+    if isinstance(response_format, NotGiven) or response_format is None:  # pyright: ignore[reportUnnecessaryComparison]
+        return Transcription
+
+    if response_format == "json":
+        return Transcription
+    elif response_format == "verbose_json":
+        return TranscriptionVerbose
+    elif response_format == "srt" or response_format == "text" or response_format == "vtt":
+        return str
+    elif TYPE_CHECKING:  # type: ignore[unreachable]
+        assert_never(response_format)
+    else:
+        log.warn("Unexpected audio response format: %s", response_format)
+        return Transcription

backend/venv/lib/python3.12/site-packages/openai/resources/audio/translations.py

@@ -0,0 +1,373 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+import logging
+from typing import TYPE_CHECKING, Union, Mapping, cast
+from typing_extensions import Literal, overload, assert_never
+
+import httpx
+
+from ... import _legacy_response
+from ...types import AudioResponseFormat
+from ..._types import NOT_GIVEN, Body, Query, Headers, NotGiven, FileTypes
+from ..._utils import (
+    extract_files,
+    maybe_transform,
+    deepcopy_minimal,
+    async_maybe_transform,
+)
+from ..._compat import cached_property
+from ..._resource import SyncAPIResource, AsyncAPIResource
+from ..._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
+from ...types.audio import translation_create_params
+from ..._base_client import make_request_options
+from ...types.audio_model import AudioModel
+from ...types.audio.translation import Translation
+from ...types.audio_response_format import AudioResponseFormat
+from ...types.audio.translation_verbose import TranslationVerbose
+
+__all__ = ["Translations", "AsyncTranslations"]
+
+log: logging.Logger = logging.getLogger("openai.audio.transcriptions")
+
+
+class Translations(SyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> TranslationsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return TranslationsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> TranslationsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return TranslationsWithStreamingResponse(self)
+
+    @overload
+    def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        response_format: Union[Literal["json"], NotGiven] = NOT_GIVEN,
+        prompt: str | NotGiven = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Translation: ...
+
+    @overload
+    def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        response_format: Literal["verbose_json"],
+        prompt: str | NotGiven = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> TranslationVerbose: ...
+
+    @overload
+    def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        response_format: Literal["text", "srt", "vtt"],
+        prompt: str | NotGiven = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> str: ...
+
+    def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        prompt: str | NotGiven = NOT_GIVEN,
+        response_format: Union[AudioResponseFormat, NotGiven] = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Translation | TranslationVerbose | str:
+        """
+        Translates audio into English.
+
+        Args:
+          file: The audio file object (not file name) translate, in one of these formats: flac,
+              mp3, mp4, mpeg, mpga, m4a, ogg, wav, or webm.
+
+          model: ID of the model to use. Only `whisper-1` (which is powered by our open source
+              Whisper V2 model) is currently available.
+
+          prompt: An optional text to guide the model's style or continue a previous audio
+              segment. The
+              [prompt](https://platform.openai.com/docs/guides/speech-to-text/prompting)
+              should be in English.
+
+          response_format: The format of the output, in one of these options: `json`, `text`, `srt`,
+              `verbose_json`, or `vtt`.
+
+          temperature: The sampling temperature, between 0 and 1. Higher values like 0.8 will make the
+              output more random, while lower values like 0.2 will make it more focused and
+              deterministic. If set to 0, the model will use
+              [log probability](https://en.wikipedia.org/wiki/Log_probability) to
+              automatically increase the temperature until certain thresholds are hit.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        body = deepcopy_minimal(
+            {
+                "file": file,
+                "model": model,
+                "prompt": prompt,
+                "response_format": response_format,
+                "temperature": temperature,
+            }
+        )
+        files = extract_files(cast(Mapping[str, object], body), paths=[["file"]])
+        # It should be noted that the actual Content-Type header that will be
+        # sent to the server will contain a `boundary` parameter, e.g.
+        # multipart/form-data; boundary=---abc--
+        extra_headers = {"Content-Type": "multipart/form-data", **(extra_headers or {})}
+        return self._post(  # type: ignore[return-value]
+            "/audio/translations",
+            body=maybe_transform(body, translation_create_params.TranslationCreateParams),
+            files=files,
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=_get_response_format_type(response_format),
+        )
+
+
+class AsyncTranslations(AsyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> AsyncTranslationsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncTranslationsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncTranslationsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncTranslationsWithStreamingResponse(self)
+
+    @overload
+    async def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        response_format: Union[Literal["json"], NotGiven] = NOT_GIVEN,
+        prompt: str | NotGiven = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Translation: ...
+
+    @overload
+    async def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        response_format: Literal["verbose_json"],
+        prompt: str | NotGiven = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> TranslationVerbose: ...
+
+    @overload
+    async def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        response_format: Literal["text", "srt", "vtt"],
+        prompt: str | NotGiven = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> str: ...
+
+    async def create(
+        self,
+        *,
+        file: FileTypes,
+        model: Union[str, AudioModel],
+        prompt: str | NotGiven = NOT_GIVEN,
+        response_format: Union[AudioResponseFormat, NotGiven] = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Translation | TranslationVerbose | str:
+        """
+        Translates audio into English.
+
+        Args:
+          file: The audio file object (not file name) translate, in one of these formats: flac,
+              mp3, mp4, mpeg, mpga, m4a, ogg, wav, or webm.
+
+          model: ID of the model to use. Only `whisper-1` (which is powered by our open source
+              Whisper V2 model) is currently available.
+
+          prompt: An optional text to guide the model's style or continue a previous audio
+              segment. The
+              [prompt](https://platform.openai.com/docs/guides/speech-to-text/prompting)
+              should be in English.
+
+          response_format: The format of the output, in one of these options: `json`, `text`, `srt`,
+              `verbose_json`, or `vtt`.
+
+          temperature: The sampling temperature, between 0 and 1. Higher values like 0.8 will make the
+              output more random, while lower values like 0.2 will make it more focused and
+              deterministic. If set to 0, the model will use
+              [log probability](https://en.wikipedia.org/wiki/Log_probability) to
+              automatically increase the temperature until certain thresholds are hit.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        body = deepcopy_minimal(
+            {
+                "file": file,
+                "model": model,
+                "prompt": prompt,
+                "response_format": response_format,
+                "temperature": temperature,
+            }
+        )
+        files = extract_files(cast(Mapping[str, object], body), paths=[["file"]])
+        # It should be noted that the actual Content-Type header that will be
+        # sent to the server will contain a `boundary` parameter, e.g.
+        # multipart/form-data; boundary=---abc--
+        extra_headers = {"Content-Type": "multipart/form-data", **(extra_headers or {})}
+        return await self._post(
+            "/audio/translations",
+            body=await async_maybe_transform(body, translation_create_params.TranslationCreateParams),
+            files=files,
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=_get_response_format_type(response_format),
+        )
+
+
+class TranslationsWithRawResponse:
+    def __init__(self, translations: Translations) -> None:
+        self._translations = translations
+
+        self.create = _legacy_response.to_raw_response_wrapper(
+            translations.create,
+        )
+
+
+class AsyncTranslationsWithRawResponse:
+    def __init__(self, translations: AsyncTranslations) -> None:
+        self._translations = translations
+
+        self.create = _legacy_response.async_to_raw_response_wrapper(
+            translations.create,
+        )
+
+
+class TranslationsWithStreamingResponse:
+    def __init__(self, translations: Translations) -> None:
+        self._translations = translations
+
+        self.create = to_streamed_response_wrapper(
+            translations.create,
+        )
+
+
+class AsyncTranslationsWithStreamingResponse:
+    def __init__(self, translations: AsyncTranslations) -> None:
+        self._translations = translations
+
+        self.create = async_to_streamed_response_wrapper(
+            translations.create,
+        )
+
+
+def _get_response_format_type(
+    response_format: Literal["json", "text", "srt", "verbose_json", "vtt"] | NotGiven,
+) -> type[Translation | TranslationVerbose | str]:
+    if isinstance(response_format, NotGiven) or response_format is None:  # pyright: ignore[reportUnnecessaryComparison]
+        return Translation
+
+    if response_format == "json":
+        return Translation
+    elif response_format == "verbose_json":
+        return TranslationVerbose
+    elif response_format == "srt" or response_format == "text" or response_format == "vtt":
+        return str
+    elif TYPE_CHECKING:  # type: ignore[unreachable]
+        assert_never(response_format)
+    else:
+        log.warn("Unexpected audio response format: %s", response_format)
+        return Transcription

backend/venv/lib/python3.12/site-packages/openai/resources/batches.py

@@ -0,0 +1,509 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Dict, Optional
+from typing_extensions import Literal
+
+import httpx
+
+from .. import _legacy_response
+from ..types import batch_list_params, batch_create_params
+from .._types import NOT_GIVEN, Body, Query, Headers, NotGiven
+from .._utils import (
+    maybe_transform,
+    async_maybe_transform,
+)
+from .._compat import cached_property
+from .._resource import SyncAPIResource, AsyncAPIResource
+from .._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
+from ..pagination import SyncCursorPage, AsyncCursorPage
+from ..types.batch import Batch
+from .._base_client import (
+    AsyncPaginator,
+    make_request_options,
+)
+
+__all__ = ["Batches", "AsyncBatches"]
+
+
+class Batches(SyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> BatchesWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return BatchesWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> BatchesWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return BatchesWithStreamingResponse(self)
+
+    def create(
+        self,
+        *,
+        completion_window: Literal["24h"],
+        endpoint: Literal["/v1/chat/completions", "/v1/embeddings", "/v1/completions"],
+        input_file_id: str,
+        metadata: Optional[Dict[str, str]] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Batch:
+        """
+        Creates and executes a batch from an uploaded file of requests
+
+        Args:
+          completion_window: The time frame within which the batch should be processed. Currently only `24h`
+              is supported.
+
+          endpoint: The endpoint to be used for all requests in the batch. Currently
+              `/v1/chat/completions`, `/v1/embeddings`, and `/v1/completions` are supported.
+              Note that `/v1/embeddings` batches are also restricted to a maximum of 50,000
+              embedding inputs across all requests in the batch.
+
+          input_file_id: The ID of an uploaded file that contains requests for the new batch.
+
+              See [upload file](https://platform.openai.com/docs/api-reference/files/create)
+              for how to upload a file.
+
+              Your input file must be formatted as a
+              [JSONL file](https://platform.openai.com/docs/api-reference/batch/request-input),
+              and must be uploaded with the purpose `batch`. The file can contain up to 50,000
+              requests, and can be up to 100 MB in size.
+
+          metadata: Optional custom metadata for the batch.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        return self._post(
+            "/batches",
+            body=maybe_transform(
+                {
+                    "completion_window": completion_window,
+                    "endpoint": endpoint,
+                    "input_file_id": input_file_id,
+                    "metadata": metadata,
+                },
+                batch_create_params.BatchCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=Batch,
+        )
+
+    def retrieve(
+        self,
+        batch_id: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Batch:
+        """
+        Retrieves a batch.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not batch_id:
+            raise ValueError(f"Expected a non-empty value for `batch_id` but received {batch_id!r}")
+        return self._get(
+            f"/batches/{batch_id}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=Batch,
+        )
+
+    def list(
+        self,
+        *,
+        after: str | NotGiven = NOT_GIVEN,
+        limit: int | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> SyncCursorPage[Batch]:
+        """List your organization's batches.
+
+        Args:
+          after: A cursor for use in pagination.
+
+        `after` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include after=obj_foo in order to
+              fetch the next page of the list.
+
+          limit: A limit on the number of objects to be returned. Limit can range between 1 and
+              100, and the default is 20.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        return self._get_api_list(
+            "/batches",
+            page=SyncCursorPage[Batch],
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform(
+                    {
+                        "after": after,
+                        "limit": limit,
+                    },
+                    batch_list_params.BatchListParams,
+                ),
+            ),
+            model=Batch,
+        )
+
+    def cancel(
+        self,
+        batch_id: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Batch:
+        """Cancels an in-progress batch.
+
+        The batch will be in status `cancelling` for up to
+        10 minutes, before changing to `cancelled`, where it will have partial results
+        (if any) available in the output file.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not batch_id:
+            raise ValueError(f"Expected a non-empty value for `batch_id` but received {batch_id!r}")
+        return self._post(
+            f"/batches/{batch_id}/cancel",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=Batch,
+        )
+
+
+class AsyncBatches(AsyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> AsyncBatchesWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncBatchesWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncBatchesWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncBatchesWithStreamingResponse(self)
+
+    async def create(
+        self,
+        *,
+        completion_window: Literal["24h"],
+        endpoint: Literal["/v1/chat/completions", "/v1/embeddings", "/v1/completions"],
+        input_file_id: str,
+        metadata: Optional[Dict[str, str]] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Batch:
+        """
+        Creates and executes a batch from an uploaded file of requests
+
+        Args:
+          completion_window: The time frame within which the batch should be processed. Currently only `24h`
+              is supported.
+
+          endpoint: The endpoint to be used for all requests in the batch. Currently
+              `/v1/chat/completions`, `/v1/embeddings`, and `/v1/completions` are supported.
+              Note that `/v1/embeddings` batches are also restricted to a maximum of 50,000
+              embedding inputs across all requests in the batch.
+
+          input_file_id: The ID of an uploaded file that contains requests for the new batch.
+
+              See [upload file](https://platform.openai.com/docs/api-reference/files/create)
+              for how to upload a file.
+
+              Your input file must be formatted as a
+              [JSONL file](https://platform.openai.com/docs/api-reference/batch/request-input),
+              and must be uploaded with the purpose `batch`. The file can contain up to 50,000
+              requests, and can be up to 100 MB in size.
+
+          metadata: Optional custom metadata for the batch.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        return await self._post(
+            "/batches",
+            body=await async_maybe_transform(
+                {
+                    "completion_window": completion_window,
+                    "endpoint": endpoint,
+                    "input_file_id": input_file_id,
+                    "metadata": metadata,
+                },
+                batch_create_params.BatchCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=Batch,
+        )
+
+    async def retrieve(
+        self,
+        batch_id: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Batch:
+        """
+        Retrieves a batch.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not batch_id:
+            raise ValueError(f"Expected a non-empty value for `batch_id` but received {batch_id!r}")
+        return await self._get(
+            f"/batches/{batch_id}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=Batch,
+        )
+
+    def list(
+        self,
+        *,
+        after: str | NotGiven = NOT_GIVEN,
+        limit: int | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> AsyncPaginator[Batch, AsyncCursorPage[Batch]]:
+        """List your organization's batches.
+
+        Args:
+          after: A cursor for use in pagination.
+
+        `after` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include after=obj_foo in order to
+              fetch the next page of the list.
+
+          limit: A limit on the number of objects to be returned. Limit can range between 1 and
+              100, and the default is 20.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        return self._get_api_list(
+            "/batches",
+            page=AsyncCursorPage[Batch],
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform(
+                    {
+                        "after": after,
+                        "limit": limit,
+                    },
+                    batch_list_params.BatchListParams,
+                ),
+            ),
+            model=Batch,
+        )
+
+    async def cancel(
+        self,
+        batch_id: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Batch:
+        """Cancels an in-progress batch.
+
+        The batch will be in status `cancelling` for up to
+        10 minutes, before changing to `cancelled`, where it will have partial results
+        (if any) available in the output file.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not batch_id:
+            raise ValueError(f"Expected a non-empty value for `batch_id` but received {batch_id!r}")
+        return await self._post(
+            f"/batches/{batch_id}/cancel",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=Batch,
+        )
+
+
+class BatchesWithRawResponse:
+    def __init__(self, batches: Batches) -> None:
+        self._batches = batches
+
+        self.create = _legacy_response.to_raw_response_wrapper(
+            batches.create,
+        )
+        self.retrieve = _legacy_response.to_raw_response_wrapper(
+            batches.retrieve,
+        )
+        self.list = _legacy_response.to_raw_response_wrapper(
+            batches.list,
+        )
+        self.cancel = _legacy_response.to_raw_response_wrapper(
+            batches.cancel,
+        )
+
+
+class AsyncBatchesWithRawResponse:
+    def __init__(self, batches: AsyncBatches) -> None:
+        self._batches = batches
+
+        self.create = _legacy_response.async_to_raw_response_wrapper(
+            batches.create,
+        )
+        self.retrieve = _legacy_response.async_to_raw_response_wrapper(
+            batches.retrieve,
+        )
+        self.list = _legacy_response.async_to_raw_response_wrapper(
+            batches.list,
+        )
+        self.cancel = _legacy_response.async_to_raw_response_wrapper(
+            batches.cancel,
+        )
+
+
+class BatchesWithStreamingResponse:
+    def __init__(self, batches: Batches) -> None:
+        self._batches = batches
+
+        self.create = to_streamed_response_wrapper(
+            batches.create,
+        )
+        self.retrieve = to_streamed_response_wrapper(
+            batches.retrieve,
+        )
+        self.list = to_streamed_response_wrapper(
+            batches.list,
+        )
+        self.cancel = to_streamed_response_wrapper(
+            batches.cancel,
+        )
+
+
+class AsyncBatchesWithStreamingResponse:
+    def __init__(self, batches: AsyncBatches) -> None:
+        self._batches = batches
+
+        self.create = async_to_streamed_response_wrapper(
+            batches.create,
+        )
+        self.retrieve = async_to_streamed_response_wrapper(
+            batches.retrieve,
+        )
+        self.list = async_to_streamed_response_wrapper(
+            batches.list,
+        )
+        self.cancel = async_to_streamed_response_wrapper(
+            batches.cancel,
+        )

backend/venv/lib/python3.12/site-packages/openai/resources/beta/__init__.py

@@ -0,0 +1,61 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from .beta import (
+    Beta,
+    AsyncBeta,
+    BetaWithRawResponse,
+    AsyncBetaWithRawResponse,
+    BetaWithStreamingResponse,
+    AsyncBetaWithStreamingResponse,
+)
+from .threads import (
+    Threads,
+    AsyncThreads,
+    ThreadsWithRawResponse,
+    AsyncThreadsWithRawResponse,
+    ThreadsWithStreamingResponse,
+    AsyncThreadsWithStreamingResponse,
+)
+from .assistants import (
+    Assistants,
+    AsyncAssistants,
+    AssistantsWithRawResponse,
+    AsyncAssistantsWithRawResponse,
+    AssistantsWithStreamingResponse,
+    AsyncAssistantsWithStreamingResponse,
+)
+from .vector_stores import (
+    VectorStores,
+    AsyncVectorStores,
+    VectorStoresWithRawResponse,
+    AsyncVectorStoresWithRawResponse,
+    VectorStoresWithStreamingResponse,
+    AsyncVectorStoresWithStreamingResponse,
+)
+
+__all__ = [
+    "VectorStores",
+    "AsyncVectorStores",
+    "VectorStoresWithRawResponse",
+    "AsyncVectorStoresWithRawResponse",
+    "VectorStoresWithStreamingResponse",
+    "AsyncVectorStoresWithStreamingResponse",
+    "Assistants",
+    "AsyncAssistants",
+    "AssistantsWithRawResponse",
+    "AsyncAssistantsWithRawResponse",
+    "AssistantsWithStreamingResponse",
+    "AsyncAssistantsWithStreamingResponse",
+    "Threads",
+    "AsyncThreads",
+    "ThreadsWithRawResponse",
+    "AsyncThreadsWithRawResponse",
+    "ThreadsWithStreamingResponse",
+    "AsyncThreadsWithStreamingResponse",
+    "Beta",
+    "AsyncBeta",
+    "BetaWithRawResponse",
+    "AsyncBetaWithRawResponse",
+    "BetaWithStreamingResponse",
+    "AsyncBetaWithStreamingResponse",
+]

backend/venv/lib/python3.12/site-packages/openai/resources/beta/assistants.py

@@ -0,0 +1,888 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Union, Iterable, Optional
+from typing_extensions import Literal
+
+import httpx
+
+from ... import _legacy_response
+from ..._types import NOT_GIVEN, Body, Query, Headers, NotGiven
+from ..._utils import (
+    maybe_transform,
+    async_maybe_transform,
+)
+from ..._compat import cached_property
+from ..._resource import SyncAPIResource, AsyncAPIResource
+from ..._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
+from ...pagination import SyncCursorPage, AsyncCursorPage
+from ...types.beta import (
+    assistant_list_params,
+    assistant_create_params,
+    assistant_update_params,
+)
+from ..._base_client import AsyncPaginator, make_request_options
+from ...types.chat_model import ChatModel
+from ...types.beta.assistant import Assistant
+from ...types.beta.assistant_deleted import AssistantDeleted
+from ...types.beta.assistant_tool_param import AssistantToolParam
+from ...types.beta.assistant_response_format_option_param import AssistantResponseFormatOptionParam
+
+__all__ = ["Assistants", "AsyncAssistants"]
+
+
+class Assistants(SyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> AssistantsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AssistantsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AssistantsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AssistantsWithStreamingResponse(self)
+
+    def create(
+        self,
+        *,
+        model: Union[str, ChatModel],
+        description: Optional[str] | NotGiven = NOT_GIVEN,
+        instructions: Optional[str] | NotGiven = NOT_GIVEN,
+        metadata: Optional[object] | NotGiven = NOT_GIVEN,
+        name: Optional[str] | NotGiven = NOT_GIVEN,
+        response_format: Optional[AssistantResponseFormatOptionParam] | NotGiven = NOT_GIVEN,
+        temperature: Optional[float] | NotGiven = NOT_GIVEN,
+        tool_resources: Optional[assistant_create_params.ToolResources] | NotGiven = NOT_GIVEN,
+        tools: Iterable[AssistantToolParam] | NotGiven = NOT_GIVEN,
+        top_p: Optional[float] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Assistant:
+        """
+        Create an assistant with a model and instructions.
+
+        Args:
+          model: ID of the model to use. You can use the
+              [List models](https://platform.openai.com/docs/api-reference/models/list) API to
+              see all of your available models, or see our
+              [Model overview](https://platform.openai.com/docs/models/overview) for
+              descriptions of them.
+
+          description: The description of the assistant. The maximum length is 512 characters.
+
+          instructions: The system instructions that the assistant uses. The maximum length is 256,000
+              characters.
+
+          metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
+              for storing additional information about the object in a structured format. Keys
+              can be a maximum of 64 characters long and values can be a maximum of 512
+              characters long.
+
+          name: The name of the assistant. The maximum length is 256 characters.
+
+          response_format: Specifies the format that the model must output. Compatible with
+              [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
+              [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4),
+              and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.
+
+              Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
+              Outputs which ensures the model will match your supplied JSON schema. Learn more
+              in the
+              [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
+
+              Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
+              message the model generates is valid JSON.
+
+              **Important:** when using JSON mode, you **must** also instruct the model to
+              produce JSON yourself via a system or user message. Without this, the model may
+              generate an unending stream of whitespace until the generation reaches the token
+              limit, resulting in a long-running and seemingly "stuck" request. Also note that
+              the message content may be partially cut off if `finish_reason="length"`, which
+              indicates the generation exceeded `max_tokens` or the conversation exceeded the
+              max context length.
+
+          temperature: What sampling temperature to use, between 0 and 2. Higher values like 0.8 will
+              make the output more random, while lower values like 0.2 will make it more
+              focused and deterministic.
+
+          tool_resources: A set of resources that are used by the assistant's tools. The resources are
+              specific to the type of tool. For example, the `code_interpreter` tool requires
+              a list of file IDs, while the `file_search` tool requires a list of vector store
+              IDs.
+
+          tools: A list of tool enabled on the assistant. There can be a maximum of 128 tools per
+              assistant. Tools can be of types `code_interpreter`, `file_search`, or
+              `function`.
+
+          top_p: An alternative to sampling with temperature, called nucleus sampling, where the
+              model considers the results of the tokens with top_p probability mass. So 0.1
+              means only the tokens comprising the top 10% probability mass are considered.
+
+              We generally recommend altering this or temperature but not both.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._post(
+            "/assistants",
+            body=maybe_transform(
+                {
+                    "model": model,
+                    "description": description,
+                    "instructions": instructions,
+                    "metadata": metadata,
+                    "name": name,
+                    "response_format": response_format,
+                    "temperature": temperature,
+                    "tool_resources": tool_resources,
+                    "tools": tools,
+                    "top_p": top_p,
+                },
+                assistant_create_params.AssistantCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=Assistant,
+        )
+
+    def retrieve(
+        self,
+        assistant_id: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Assistant:
+        """
+        Retrieves an assistant.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not assistant_id:
+            raise ValueError(f"Expected a non-empty value for `assistant_id` but received {assistant_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._get(
+            f"/assistants/{assistant_id}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=Assistant,
+        )
+
+    def update(
+        self,
+        assistant_id: str,
+        *,
+        description: Optional[str] | NotGiven = NOT_GIVEN,
+        instructions: Optional[str] | NotGiven = NOT_GIVEN,
+        metadata: Optional[object] | NotGiven = NOT_GIVEN,
+        model: str | NotGiven = NOT_GIVEN,
+        name: Optional[str] | NotGiven = NOT_GIVEN,
+        response_format: Optional[AssistantResponseFormatOptionParam] | NotGiven = NOT_GIVEN,
+        temperature: Optional[float] | NotGiven = NOT_GIVEN,
+        tool_resources: Optional[assistant_update_params.ToolResources] | NotGiven = NOT_GIVEN,
+        tools: Iterable[AssistantToolParam] | NotGiven = NOT_GIVEN,
+        top_p: Optional[float] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Assistant:
+        """Modifies an assistant.
+
+        Args:
+          description: The description of the assistant.
+
+        The maximum length is 512 characters.
+
+          instructions: The system instructions that the assistant uses. The maximum length is 256,000
+              characters.
+
+          metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
+              for storing additional information about the object in a structured format. Keys
+              can be a maximum of 64 characters long and values can be a maximum of 512
+              characters long.
+
+          model: ID of the model to use. You can use the
+              [List models](https://platform.openai.com/docs/api-reference/models/list) API to
+              see all of your available models, or see our
+              [Model overview](https://platform.openai.com/docs/models/overview) for
+              descriptions of them.
+
+          name: The name of the assistant. The maximum length is 256 characters.
+
+          response_format: Specifies the format that the model must output. Compatible with
+              [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
+              [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4),
+              and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.
+
+              Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
+              Outputs which ensures the model will match your supplied JSON schema. Learn more
+              in the
+              [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
+
+              Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
+              message the model generates is valid JSON.
+
+              **Important:** when using JSON mode, you **must** also instruct the model to
+              produce JSON yourself via a system or user message. Without this, the model may
+              generate an unending stream of whitespace until the generation reaches the token
+              limit, resulting in a long-running and seemingly "stuck" request. Also note that
+              the message content may be partially cut off if `finish_reason="length"`, which
+              indicates the generation exceeded `max_tokens` or the conversation exceeded the
+              max context length.
+
+          temperature: What sampling temperature to use, between 0 and 2. Higher values like 0.8 will
+              make the output more random, while lower values like 0.2 will make it more
+              focused and deterministic.
+
+          tool_resources: A set of resources that are used by the assistant's tools. The resources are
+              specific to the type of tool. For example, the `code_interpreter` tool requires
+              a list of file IDs, while the `file_search` tool requires a list of vector store
+              IDs.
+
+          tools: A list of tool enabled on the assistant. There can be a maximum of 128 tools per
+              assistant. Tools can be of types `code_interpreter`, `file_search`, or
+              `function`.
+
+          top_p: An alternative to sampling with temperature, called nucleus sampling, where the
+              model considers the results of the tokens with top_p probability mass. So 0.1
+              means only the tokens comprising the top 10% probability mass are considered.
+
+              We generally recommend altering this or temperature but not both.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not assistant_id:
+            raise ValueError(f"Expected a non-empty value for `assistant_id` but received {assistant_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._post(
+            f"/assistants/{assistant_id}",
+            body=maybe_transform(
+                {
+                    "description": description,
+                    "instructions": instructions,
+                    "metadata": metadata,
+                    "model": model,
+                    "name": name,
+                    "response_format": response_format,
+                    "temperature": temperature,
+                    "tool_resources": tool_resources,
+                    "tools": tools,
+                    "top_p": top_p,
+                },
+                assistant_update_params.AssistantUpdateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=Assistant,
+        )
+
+    def list(
+        self,
+        *,
+        after: str | NotGiven = NOT_GIVEN,
+        before: str | NotGiven = NOT_GIVEN,
+        limit: int | NotGiven = NOT_GIVEN,
+        order: Literal["asc", "desc"] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> SyncCursorPage[Assistant]:
+        """Returns a list of assistants.
+
+        Args:
+          after: A cursor for use in pagination.
+
+        `after` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include after=obj_foo in order to
+              fetch the next page of the list.
+
+          before: A cursor for use in pagination. `before` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include before=obj_foo in order to
+              fetch the previous page of the list.
+
+          limit: A limit on the number of objects to be returned. Limit can range between 1 and
+              100, and the default is 20.
+
+          order: Sort order by the `created_at` timestamp of the objects. `asc` for ascending
+              order and `desc` for descending order.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._get_api_list(
+            "/assistants",
+            page=SyncCursorPage[Assistant],
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform(
+                    {
+                        "after": after,
+                        "before": before,
+                        "limit": limit,
+                        "order": order,
+                    },
+                    assistant_list_params.AssistantListParams,
+                ),
+            ),
+            model=Assistant,
+        )
+
+    def delete(
+        self,
+        assistant_id: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> AssistantDeleted:
+        """
+        Delete an assistant.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not assistant_id:
+            raise ValueError(f"Expected a non-empty value for `assistant_id` but received {assistant_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._delete(
+            f"/assistants/{assistant_id}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=AssistantDeleted,
+        )
+
+
+class AsyncAssistants(AsyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> AsyncAssistantsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncAssistantsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncAssistantsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncAssistantsWithStreamingResponse(self)
+
+    async def create(
+        self,
+        *,
+        model: Union[str, ChatModel],
+        description: Optional[str] | NotGiven = NOT_GIVEN,
+        instructions: Optional[str] | NotGiven = NOT_GIVEN,
+        metadata: Optional[object] | NotGiven = NOT_GIVEN,
+        name: Optional[str] | NotGiven = NOT_GIVEN,
+        response_format: Optional[AssistantResponseFormatOptionParam] | NotGiven = NOT_GIVEN,
+        temperature: Optional[float] | NotGiven = NOT_GIVEN,
+        tool_resources: Optional[assistant_create_params.ToolResources] | NotGiven = NOT_GIVEN,
+        tools: Iterable[AssistantToolParam] | NotGiven = NOT_GIVEN,
+        top_p: Optional[float] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Assistant:
+        """
+        Create an assistant with a model and instructions.
+
+        Args:
+          model: ID of the model to use. You can use the
+              [List models](https://platform.openai.com/docs/api-reference/models/list) API to
+              see all of your available models, or see our
+              [Model overview](https://platform.openai.com/docs/models/overview) for
+              descriptions of them.
+
+          description: The description of the assistant. The maximum length is 512 characters.
+
+          instructions: The system instructions that the assistant uses. The maximum length is 256,000
+              characters.
+
+          metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
+              for storing additional information about the object in a structured format. Keys
+              can be a maximum of 64 characters long and values can be a maximum of 512
+              characters long.
+
+          name: The name of the assistant. The maximum length is 256 characters.
+
+          response_format: Specifies the format that the model must output. Compatible with
+              [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
+              [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4),
+              and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.
+
+              Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
+              Outputs which ensures the model will match your supplied JSON schema. Learn more
+              in the
+              [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
+
+              Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
+              message the model generates is valid JSON.
+
+              **Important:** when using JSON mode, you **must** also instruct the model to
+              produce JSON yourself via a system or user message. Without this, the model may
+              generate an unending stream of whitespace until the generation reaches the token
+              limit, resulting in a long-running and seemingly "stuck" request. Also note that
+              the message content may be partially cut off if `finish_reason="length"`, which
+              indicates the generation exceeded `max_tokens` or the conversation exceeded the
+              max context length.
+
+          temperature: What sampling temperature to use, between 0 and 2. Higher values like 0.8 will
+              make the output more random, while lower values like 0.2 will make it more
+              focused and deterministic.
+
+          tool_resources: A set of resources that are used by the assistant's tools. The resources are
+              specific to the type of tool. For example, the `code_interpreter` tool requires
+              a list of file IDs, while the `file_search` tool requires a list of vector store
+              IDs.
+
+          tools: A list of tool enabled on the assistant. There can be a maximum of 128 tools per
+              assistant. Tools can be of types `code_interpreter`, `file_search`, or
+              `function`.
+
+          top_p: An alternative to sampling with temperature, called nucleus sampling, where the
+              model considers the results of the tokens with top_p probability mass. So 0.1
+              means only the tokens comprising the top 10% probability mass are considered.
+
+              We generally recommend altering this or temperature but not both.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return await self._post(
+            "/assistants",
+            body=await async_maybe_transform(
+                {
+                    "model": model,
+                    "description": description,
+                    "instructions": instructions,
+                    "metadata": metadata,
+                    "name": name,
+                    "response_format": response_format,
+                    "temperature": temperature,
+                    "tool_resources": tool_resources,
+                    "tools": tools,
+                    "top_p": top_p,
+                },
+                assistant_create_params.AssistantCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=Assistant,
+        )
+
+    async def retrieve(
+        self,
+        assistant_id: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Assistant:
+        """
+        Retrieves an assistant.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not assistant_id:
+            raise ValueError(f"Expected a non-empty value for `assistant_id` but received {assistant_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return await self._get(
+            f"/assistants/{assistant_id}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=Assistant,
+        )
+
+    async def update(
+        self,
+        assistant_id: str,
+        *,
+        description: Optional[str] | NotGiven = NOT_GIVEN,
+        instructions: Optional[str] | NotGiven = NOT_GIVEN,
+        metadata: Optional[object] | NotGiven = NOT_GIVEN,
+        model: str | NotGiven = NOT_GIVEN,
+        name: Optional[str] | NotGiven = NOT_GIVEN,
+        response_format: Optional[AssistantResponseFormatOptionParam] | NotGiven = NOT_GIVEN,
+        temperature: Optional[float] | NotGiven = NOT_GIVEN,
+        tool_resources: Optional[assistant_update_params.ToolResources] | NotGiven = NOT_GIVEN,
+        tools: Iterable[AssistantToolParam] | NotGiven = NOT_GIVEN,
+        top_p: Optional[float] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Assistant:
+        """Modifies an assistant.
+
+        Args:
+          description: The description of the assistant.
+
+        The maximum length is 512 characters.
+
+          instructions: The system instructions that the assistant uses. The maximum length is 256,000
+              characters.
+
+          metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
+              for storing additional information about the object in a structured format. Keys
+              can be a maximum of 64 characters long and values can be a maximum of 512
+              characters long.
+
+          model: ID of the model to use. You can use the
+              [List models](https://platform.openai.com/docs/api-reference/models/list) API to
+              see all of your available models, or see our
+              [Model overview](https://platform.openai.com/docs/models/overview) for
+              descriptions of them.
+
+          name: The name of the assistant. The maximum length is 256 characters.
+
+          response_format: Specifies the format that the model must output. Compatible with
+              [GPT-4o](https://platform.openai.com/docs/models/gpt-4o),
+              [GPT-4 Turbo](https://platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4),
+              and all GPT-3.5 Turbo models since `gpt-3.5-turbo-1106`.
+
+              Setting to `{ "type": "json_schema", "json_schema": {...} }` enables Structured
+              Outputs which ensures the model will match your supplied JSON schema. Learn more
+              in the
+              [Structured Outputs guide](https://platform.openai.com/docs/guides/structured-outputs).
+
+              Setting to `{ "type": "json_object" }` enables JSON mode, which ensures the
+              message the model generates is valid JSON.
+
+              **Important:** when using JSON mode, you **must** also instruct the model to
+              produce JSON yourself via a system or user message. Without this, the model may
+              generate an unending stream of whitespace until the generation reaches the token
+              limit, resulting in a long-running and seemingly "stuck" request. Also note that
+              the message content may be partially cut off if `finish_reason="length"`, which
+              indicates the generation exceeded `max_tokens` or the conversation exceeded the
+              max context length.
+
+          temperature: What sampling temperature to use, between 0 and 2. Higher values like 0.8 will
+              make the output more random, while lower values like 0.2 will make it more
+              focused and deterministic.
+
+          tool_resources: A set of resources that are used by the assistant's tools. The resources are
+              specific to the type of tool. For example, the `code_interpreter` tool requires
+              a list of file IDs, while the `file_search` tool requires a list of vector store
+              IDs.
+
+          tools: A list of tool enabled on the assistant. There can be a maximum of 128 tools per
+              assistant. Tools can be of types `code_interpreter`, `file_search`, or
+              `function`.
+
+          top_p: An alternative to sampling with temperature, called nucleus sampling, where the
+              model considers the results of the tokens with top_p probability mass. So 0.1
+              means only the tokens comprising the top 10% probability mass are considered.
+
+              We generally recommend altering this or temperature but not both.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not assistant_id:
+            raise ValueError(f"Expected a non-empty value for `assistant_id` but received {assistant_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return await self._post(
+            f"/assistants/{assistant_id}",
+            body=await async_maybe_transform(
+                {
+                    "description": description,
+                    "instructions": instructions,
+                    "metadata": metadata,
+                    "model": model,
+                    "name": name,
+                    "response_format": response_format,
+                    "temperature": temperature,
+                    "tool_resources": tool_resources,
+                    "tools": tools,
+                    "top_p": top_p,
+                },
+                assistant_update_params.AssistantUpdateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=Assistant,
+        )
+
+    def list(
+        self,
+        *,
+        after: str | NotGiven = NOT_GIVEN,
+        before: str | NotGiven = NOT_GIVEN,
+        limit: int | NotGiven = NOT_GIVEN,
+        order: Literal["asc", "desc"] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> AsyncPaginator[Assistant, AsyncCursorPage[Assistant]]:
+        """Returns a list of assistants.
+
+        Args:
+          after: A cursor for use in pagination.
+
+        `after` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include after=obj_foo in order to
+              fetch the next page of the list.
+
+          before: A cursor for use in pagination. `before` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include before=obj_foo in order to
+              fetch the previous page of the list.
+
+          limit: A limit on the number of objects to be returned. Limit can range between 1 and
+              100, and the default is 20.
+
+          order: Sort order by the `created_at` timestamp of the objects. `asc` for ascending
+              order and `desc` for descending order.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._get_api_list(
+            "/assistants",
+            page=AsyncCursorPage[Assistant],
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform(
+                    {
+                        "after": after,
+                        "before": before,
+                        "limit": limit,
+                        "order": order,
+                    },
+                    assistant_list_params.AssistantListParams,
+                ),
+            ),
+            model=Assistant,
+        )
+
+    async def delete(
+        self,
+        assistant_id: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> AssistantDeleted:
+        """
+        Delete an assistant.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not assistant_id:
+            raise ValueError(f"Expected a non-empty value for `assistant_id` but received {assistant_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return await self._delete(
+            f"/assistants/{assistant_id}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=AssistantDeleted,
+        )
+
+
+class AssistantsWithRawResponse:
+    def __init__(self, assistants: Assistants) -> None:
+        self._assistants = assistants
+
+        self.create = _legacy_response.to_raw_response_wrapper(
+            assistants.create,
+        )
+        self.retrieve = _legacy_response.to_raw_response_wrapper(
+            assistants.retrieve,
+        )
+        self.update = _legacy_response.to_raw_response_wrapper(
+            assistants.update,
+        )
+        self.list = _legacy_response.to_raw_response_wrapper(
+            assistants.list,
+        )
+        self.delete = _legacy_response.to_raw_response_wrapper(
+            assistants.delete,
+        )
+
+
+class AsyncAssistantsWithRawResponse:
+    def __init__(self, assistants: AsyncAssistants) -> None:
+        self._assistants = assistants
+
+        self.create = _legacy_response.async_to_raw_response_wrapper(
+            assistants.create,
+        )
+        self.retrieve = _legacy_response.async_to_raw_response_wrapper(
+            assistants.retrieve,
+        )
+        self.update = _legacy_response.async_to_raw_response_wrapper(
+            assistants.update,
+        )
+        self.list = _legacy_response.async_to_raw_response_wrapper(
+            assistants.list,
+        )
+        self.delete = _legacy_response.async_to_raw_response_wrapper(
+            assistants.delete,
+        )
+
+
+class AssistantsWithStreamingResponse:
+    def __init__(self, assistants: Assistants) -> None:
+        self._assistants = assistants
+
+        self.create = to_streamed_response_wrapper(
+            assistants.create,
+        )
+        self.retrieve = to_streamed_response_wrapper(
+            assistants.retrieve,
+        )
+        self.update = to_streamed_response_wrapper(
+            assistants.update,
+        )
+        self.list = to_streamed_response_wrapper(
+            assistants.list,
+        )
+        self.delete = to_streamed_response_wrapper(
+            assistants.delete,
+        )
+
+
+class AsyncAssistantsWithStreamingResponse:
+    def __init__(self, assistants: AsyncAssistants) -> None:
+        self._assistants = assistants
+
+        self.create = async_to_streamed_response_wrapper(
+            assistants.create,
+        )
+        self.retrieve = async_to_streamed_response_wrapper(
+            assistants.retrieve,
+        )
+        self.update = async_to_streamed_response_wrapper(
+            assistants.update,
+        )
+        self.list = async_to_streamed_response_wrapper(
+            assistants.list,
+        )
+        self.delete = async_to_streamed_response_wrapper(
+            assistants.delete,
+        )

backend/venv/lib/python3.12/site-packages/openai/resources/beta/beta.py

@@ -0,0 +1,177 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from .threads import (
+    Threads,
+    AsyncThreads,
+    ThreadsWithRawResponse,
+    AsyncThreadsWithRawResponse,
+    ThreadsWithStreamingResponse,
+    AsyncThreadsWithStreamingResponse,
+)
+from ..._compat import cached_property
+from .chat.chat import Chat, AsyncChat
+from .assistants import (
+    Assistants,
+    AsyncAssistants,
+    AssistantsWithRawResponse,
+    AsyncAssistantsWithRawResponse,
+    AssistantsWithStreamingResponse,
+    AsyncAssistantsWithStreamingResponse,
+)
+from ..._resource import SyncAPIResource, AsyncAPIResource
+from .vector_stores import (
+    VectorStores,
+    AsyncVectorStores,
+    VectorStoresWithRawResponse,
+    AsyncVectorStoresWithRawResponse,
+    VectorStoresWithStreamingResponse,
+    AsyncVectorStoresWithStreamingResponse,
+)
+from .threads.threads import Threads, AsyncThreads
+from .vector_stores.vector_stores import VectorStores, AsyncVectorStores
+
+__all__ = ["Beta", "AsyncBeta"]
+
+
+class Beta(SyncAPIResource):
+    @cached_property
+    def chat(self) -> Chat:
+        return Chat(self._client)
+
+    @cached_property
+    def vector_stores(self) -> VectorStores:
+        return VectorStores(self._client)
+
+    @cached_property
+    def assistants(self) -> Assistants:
+        return Assistants(self._client)
+
+    @cached_property
+    def threads(self) -> Threads:
+        return Threads(self._client)
+
+    @cached_property
+    def with_raw_response(self) -> BetaWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return BetaWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> BetaWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return BetaWithStreamingResponse(self)
+
+
+class AsyncBeta(AsyncAPIResource):
+    @cached_property
+    def chat(self) -> AsyncChat:
+        return AsyncChat(self._client)
+
+    @cached_property
+    def vector_stores(self) -> AsyncVectorStores:
+        return AsyncVectorStores(self._client)
+
+    @cached_property
+    def assistants(self) -> AsyncAssistants:
+        return AsyncAssistants(self._client)
+
+    @cached_property
+    def threads(self) -> AsyncThreads:
+        return AsyncThreads(self._client)
+
+    @cached_property
+    def with_raw_response(self) -> AsyncBetaWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncBetaWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncBetaWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncBetaWithStreamingResponse(self)
+
+
+class BetaWithRawResponse:
+    def __init__(self, beta: Beta) -> None:
+        self._beta = beta
+
+    @cached_property
+    def vector_stores(self) -> VectorStoresWithRawResponse:
+        return VectorStoresWithRawResponse(self._beta.vector_stores)
+
+    @cached_property
+    def assistants(self) -> AssistantsWithRawResponse:
+        return AssistantsWithRawResponse(self._beta.assistants)
+
+    @cached_property
+    def threads(self) -> ThreadsWithRawResponse:
+        return ThreadsWithRawResponse(self._beta.threads)
+
+
+class AsyncBetaWithRawResponse:
+    def __init__(self, beta: AsyncBeta) -> None:
+        self._beta = beta
+
+    @cached_property
+    def vector_stores(self) -> AsyncVectorStoresWithRawResponse:
+        return AsyncVectorStoresWithRawResponse(self._beta.vector_stores)
+
+    @cached_property
+    def assistants(self) -> AsyncAssistantsWithRawResponse:
+        return AsyncAssistantsWithRawResponse(self._beta.assistants)
+
+    @cached_property
+    def threads(self) -> AsyncThreadsWithRawResponse:
+        return AsyncThreadsWithRawResponse(self._beta.threads)
+
+
+class BetaWithStreamingResponse:
+    def __init__(self, beta: Beta) -> None:
+        self._beta = beta
+
+    @cached_property
+    def vector_stores(self) -> VectorStoresWithStreamingResponse:
+        return VectorStoresWithStreamingResponse(self._beta.vector_stores)
+
+    @cached_property
+    def assistants(self) -> AssistantsWithStreamingResponse:
+        return AssistantsWithStreamingResponse(self._beta.assistants)
+
+    @cached_property
+    def threads(self) -> ThreadsWithStreamingResponse:
+        return ThreadsWithStreamingResponse(self._beta.threads)
+
+
+class AsyncBetaWithStreamingResponse:
+    def __init__(self, beta: AsyncBeta) -> None:
+        self._beta = beta
+
+    @cached_property
+    def vector_stores(self) -> AsyncVectorStoresWithStreamingResponse:
+        return AsyncVectorStoresWithStreamingResponse(self._beta.vector_stores)
+
+    @cached_property
+    def assistants(self) -> AsyncAssistantsWithStreamingResponse:
+        return AsyncAssistantsWithStreamingResponse(self._beta.assistants)
+
+    @cached_property
+    def threads(self) -> AsyncThreadsWithStreamingResponse:
+        return AsyncThreadsWithStreamingResponse(self._beta.threads)

backend/venv/lib/python3.12/site-packages/openai/resources/beta/chat/__init__.py

@@ -0,0 +1,11 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from .chat import Chat, AsyncChat
+from .completions import Completions, AsyncCompletions
+
+__all__ = [
+    "Completions",
+    "AsyncCompletions",
+    "Chat",
+    "AsyncChat",
+]

backend/venv/lib/python3.12/site-packages/openai/resources/beta/chat/chat.py

@@ -0,0 +1,21 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from ...._compat import cached_property
+from .completions import Completions, AsyncCompletions
+from ...._resource import SyncAPIResource, AsyncAPIResource
+
+__all__ = ["Chat", "AsyncChat"]
+
+
+class Chat(SyncAPIResource):
+    @cached_property
+    def completions(self) -> Completions:
+        return Completions(self._client)
+
+
+class AsyncChat(AsyncAPIResource):
+    @cached_property
+    def completions(self) -> AsyncCompletions:
+        return AsyncCompletions(self._client)

backend/venv/lib/python3.12/site-packages/openai/resources/beta/chat/completions.py

@@ -0,0 +1,591 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Dict, List, Type, Union, Iterable, Optional, cast
+from functools import partial
+from typing_extensions import Literal
+
+import httpx
+
+from .... import _legacy_response
+from ...._types import NOT_GIVEN, Body, Query, Headers, NotGiven
+from ...._utils import maybe_transform, async_maybe_transform
+from ...._compat import cached_property
+from ...._resource import SyncAPIResource, AsyncAPIResource
+from ...._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
+from ...._streaming import Stream
+from ....types.chat import completion_create_params
+from ...._base_client import make_request_options
+from ....lib._parsing import (
+    ResponseFormatT,
+    validate_input_tools as _validate_input_tools,
+    parse_chat_completion as _parse_chat_completion,
+    type_to_response_format_param as _type_to_response_format,
+)
+from ....types.chat_model import ChatModel
+from ....lib.streaming.chat import ChatCompletionStreamManager, AsyncChatCompletionStreamManager
+from ....types.chat.chat_completion import ChatCompletion
+from ....types.chat.chat_completion_chunk import ChatCompletionChunk
+from ....types.chat.parsed_chat_completion import ParsedChatCompletion
+from ....types.chat.chat_completion_tool_param import ChatCompletionToolParam
+from ....types.chat.chat_completion_message_param import ChatCompletionMessageParam
+from ....types.chat.chat_completion_stream_options_param import ChatCompletionStreamOptionsParam
+from ....types.chat.chat_completion_tool_choice_option_param import ChatCompletionToolChoiceOptionParam
+
+__all__ = ["Completions", "AsyncCompletions"]
+
+
+class Completions(SyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> CompletionsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return CompletionsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> CompletionsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return CompletionsWithStreamingResponse(self)
+
+    def parse(
+        self,
+        *,
+        messages: Iterable[ChatCompletionMessageParam],
+        model: Union[str, ChatModel],
+        response_format: type[ResponseFormatT] | NotGiven = NOT_GIVEN,
+        frequency_penalty: Optional[float] | NotGiven = NOT_GIVEN,
+        function_call: completion_create_params.FunctionCall | NotGiven = NOT_GIVEN,
+        functions: Iterable[completion_create_params.Function] | NotGiven = NOT_GIVEN,
+        logit_bias: Optional[Dict[str, int]] | NotGiven = NOT_GIVEN,
+        logprobs: Optional[bool] | NotGiven = NOT_GIVEN,
+        max_completion_tokens: Optional[int] | NotGiven = NOT_GIVEN,
+        max_tokens: Optional[int] | NotGiven = NOT_GIVEN,
+        metadata: Optional[Dict[str, str]] | NotGiven = NOT_GIVEN,
+        n: Optional[int] | NotGiven = NOT_GIVEN,
+        parallel_tool_calls: bool | NotGiven = NOT_GIVEN,
+        presence_penalty: Optional[float] | NotGiven = NOT_GIVEN,
+        seed: Optional[int] | NotGiven = NOT_GIVEN,
+        service_tier: Optional[Literal["auto", "default"]] | NotGiven = NOT_GIVEN,
+        stop: Union[Optional[str], List[str]] | NotGiven = NOT_GIVEN,
+        store: Optional[bool] | NotGiven = NOT_GIVEN,
+        stream_options: Optional[ChatCompletionStreamOptionsParam] | NotGiven = NOT_GIVEN,
+        temperature: Optional[float] | NotGiven = NOT_GIVEN,
+        tool_choice: ChatCompletionToolChoiceOptionParam | NotGiven = NOT_GIVEN,
+        tools: Iterable[ChatCompletionToolParam] | NotGiven = NOT_GIVEN,
+        top_logprobs: Optional[int] | NotGiven = NOT_GIVEN,
+        top_p: Optional[float] | NotGiven = NOT_GIVEN,
+        user: str | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> ParsedChatCompletion[ResponseFormatT]:
+        """Wrapper over the `client.chat.completions.create()` method that provides richer integrations with Python specific types
+        & returns a `ParsedChatCompletion` object, which is a subclass of the standard `ChatCompletion` class.
+
+        You can pass a pydantic model to this method and it will automatically convert the model
+        into a JSON schema, send it to the API and parse the response content back into the given model.
+
+        This method will also automatically parse `function` tool calls if:
+        - You use the `openai.pydantic_function_tool()` helper method
+        - You mark your tool schema with `"strict": True`
+
+        Example usage:
+        ```py
+        from pydantic import BaseModel
+        from openai import OpenAI
+
+
+        class Step(BaseModel):
+            explanation: str
+            output: str
+
+
+        class MathResponse(BaseModel):
+            steps: List[Step]
+            final_answer: str
+
+
+        client = OpenAI()
+        completion = client.beta.chat.completions.parse(
+            model="gpt-4o-2024-08-06",
+            messages=[
+                {"role": "system", "content": "You are a helpful math tutor."},
+                {"role": "user", "content": "solve 8x + 31 = 2"},
+            ],
+            response_format=MathResponse,
+        )
+
+        message = completion.choices[0].message
+        if message.parsed:
+            print(message.parsed.steps)
+            print("answer: ", message.parsed.final_answer)
+        ```
+        """
+        _validate_input_tools(tools)
+
+        extra_headers = {
+            "X-Stainless-Helper-Method": "beta.chat.completions.parse",
+            **(extra_headers or {}),
+        }
+
+        def parser(raw_completion: ChatCompletion) -> ParsedChatCompletion[ResponseFormatT]:
+            return _parse_chat_completion(
+                response_format=response_format,
+                chat_completion=raw_completion,
+                input_tools=tools,
+            )
+
+        return self._post(
+            "/chat/completions",
+            body=maybe_transform(
+                {
+                    "messages": messages,
+                    "model": model,
+                    "frequency_penalty": frequency_penalty,
+                    "function_call": function_call,
+                    "functions": functions,
+                    "logit_bias": logit_bias,
+                    "logprobs": logprobs,
+                    "max_completion_tokens": max_completion_tokens,
+                    "max_tokens": max_tokens,
+                    "metadata": metadata,
+                    "n": n,
+                    "parallel_tool_calls": parallel_tool_calls,
+                    "presence_penalty": presence_penalty,
+                    "response_format": _type_to_response_format(response_format),
+                    "seed": seed,
+                    "service_tier": service_tier,
+                    "stop": stop,
+                    "store": store,
+                    "stream": False,
+                    "stream_options": stream_options,
+                    "temperature": temperature,
+                    "tool_choice": tool_choice,
+                    "tools": tools,
+                    "top_logprobs": top_logprobs,
+                    "top_p": top_p,
+                    "user": user,
+                },
+                completion_create_params.CompletionCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                post_parser=parser,
+            ),
+            # we turn the `ChatCompletion` instance into a `ParsedChatCompletion`
+            # in the `parser` function above
+            cast_to=cast(Type[ParsedChatCompletion[ResponseFormatT]], ChatCompletion),
+            stream=False,
+        )
+
+    def stream(
+        self,
+        *,
+        messages: Iterable[ChatCompletionMessageParam],
+        model: Union[str, ChatModel],
+        response_format: completion_create_params.ResponseFormat | type[ResponseFormatT] | NotGiven = NOT_GIVEN,
+        frequency_penalty: Optional[float] | NotGiven = NOT_GIVEN,
+        function_call: completion_create_params.FunctionCall | NotGiven = NOT_GIVEN,
+        functions: Iterable[completion_create_params.Function] | NotGiven = NOT_GIVEN,
+        logit_bias: Optional[Dict[str, int]] | NotGiven = NOT_GIVEN,
+        logprobs: Optional[bool] | NotGiven = NOT_GIVEN,
+        max_completion_tokens: Optional[int] | NotGiven = NOT_GIVEN,
+        max_tokens: Optional[int] | NotGiven = NOT_GIVEN,
+        metadata: Optional[Dict[str, str]] | NotGiven = NOT_GIVEN,
+        n: Optional[int] | NotGiven = NOT_GIVEN,
+        parallel_tool_calls: bool | NotGiven = NOT_GIVEN,
+        presence_penalty: Optional[float] | NotGiven = NOT_GIVEN,
+        seed: Optional[int] | NotGiven = NOT_GIVEN,
+        service_tier: Optional[Literal["auto", "default"]] | NotGiven = NOT_GIVEN,
+        stop: Union[Optional[str], List[str]] | NotGiven = NOT_GIVEN,
+        store: Optional[bool] | NotGiven = NOT_GIVEN,
+        stream_options: Optional[ChatCompletionStreamOptionsParam] | NotGiven = NOT_GIVEN,
+        temperature: Optional[float] | NotGiven = NOT_GIVEN,
+        tool_choice: ChatCompletionToolChoiceOptionParam | NotGiven = NOT_GIVEN,
+        tools: Iterable[ChatCompletionToolParam] | NotGiven = NOT_GIVEN,
+        top_logprobs: Optional[int] | NotGiven = NOT_GIVEN,
+        top_p: Optional[float] | NotGiven = NOT_GIVEN,
+        user: str | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> ChatCompletionStreamManager[ResponseFormatT]:
+        """Wrapper over the `client.chat.completions.create(stream=True)` method that provides a more granular event API
+        and automatic accumulation of each delta.
+
+        This also supports all of the parsing utilities that `.parse()` does.
+
+        Unlike `.create(stream=True)`, the `.stream()` method requires usage within a context manager to prevent accidental leakage of the response:
+
+        ```py
+        with client.beta.chat.completions.stream(
+            model="gpt-4o-2024-08-06",
+            messages=[...],
+        ) as stream:
+            for event in stream:
+                if event.type == "content.delta":
+                    print(event.delta, flush=True, end="")
+        ```
+
+        When the context manager is entered, a `ChatCompletionStream` instance is returned which, like `.create(stream=True)` is an iterator. The full list of events that are yielded by the iterator are outlined in [these docs](https://github.com/openai/openai-python/blob/main/helpers.md#chat-completions-events).
+
+        When the context manager exits, the response will be closed, however the `stream` instance is still available outside
+        the context manager.
+        """
+        extra_headers = {
+            "X-Stainless-Helper-Method": "beta.chat.completions.stream",
+            **(extra_headers or {}),
+        }
+
+        api_request: partial[Stream[ChatCompletionChunk]] = partial(
+            self._client.chat.completions.create,
+            messages=messages,
+            model=model,
+            stream=True,
+            response_format=_type_to_response_format(response_format),
+            frequency_penalty=frequency_penalty,
+            function_call=function_call,
+            functions=functions,
+            logit_bias=logit_bias,
+            logprobs=logprobs,
+            max_completion_tokens=max_completion_tokens,
+            max_tokens=max_tokens,
+            metadata=metadata,
+            n=n,
+            parallel_tool_calls=parallel_tool_calls,
+            presence_penalty=presence_penalty,
+            seed=seed,
+            service_tier=service_tier,
+            store=store,
+            stop=stop,
+            stream_options=stream_options,
+            temperature=temperature,
+            tool_choice=tool_choice,
+            tools=tools,
+            top_logprobs=top_logprobs,
+            top_p=top_p,
+            user=user,
+            extra_headers=extra_headers,
+            extra_query=extra_query,
+            extra_body=extra_body,
+            timeout=timeout,
+        )
+        return ChatCompletionStreamManager(
+            api_request,
+            response_format=response_format,
+            input_tools=tools,
+        )
+
+
+class AsyncCompletions(AsyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> AsyncCompletionsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncCompletionsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncCompletionsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncCompletionsWithStreamingResponse(self)
+
+    async def parse(
+        self,
+        *,
+        messages: Iterable[ChatCompletionMessageParam],
+        model: Union[str, ChatModel],
+        response_format: type[ResponseFormatT] | NotGiven = NOT_GIVEN,
+        frequency_penalty: Optional[float] | NotGiven = NOT_GIVEN,
+        function_call: completion_create_params.FunctionCall | NotGiven = NOT_GIVEN,
+        functions: Iterable[completion_create_params.Function] | NotGiven = NOT_GIVEN,
+        logit_bias: Optional[Dict[str, int]] | NotGiven = NOT_GIVEN,
+        logprobs: Optional[bool] | NotGiven = NOT_GIVEN,
+        max_completion_tokens: Optional[int] | NotGiven = NOT_GIVEN,
+        max_tokens: Optional[int] | NotGiven = NOT_GIVEN,
+        metadata: Optional[Dict[str, str]] | NotGiven = NOT_GIVEN,
+        n: Optional[int] | NotGiven = NOT_GIVEN,
+        parallel_tool_calls: bool | NotGiven = NOT_GIVEN,
+        presence_penalty: Optional[float] | NotGiven = NOT_GIVEN,
+        seed: Optional[int] | NotGiven = NOT_GIVEN,
+        service_tier: Optional[Literal["auto", "default"]] | NotGiven = NOT_GIVEN,
+        stop: Union[Optional[str], List[str]] | NotGiven = NOT_GIVEN,
+        store: Optional[bool] | NotGiven = NOT_GIVEN,
+        stream_options: Optional[ChatCompletionStreamOptionsParam] | NotGiven = NOT_GIVEN,
+        temperature: Optional[float] | NotGiven = NOT_GIVEN,
+        tool_choice: ChatCompletionToolChoiceOptionParam | NotGiven = NOT_GIVEN,
+        tools: Iterable[ChatCompletionToolParam] | NotGiven = NOT_GIVEN,
+        top_logprobs: Optional[int] | NotGiven = NOT_GIVEN,
+        top_p: Optional[float] | NotGiven = NOT_GIVEN,
+        user: str | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> ParsedChatCompletion[ResponseFormatT]:
+        """Wrapper over the `client.chat.completions.create()` method that provides richer integrations with Python specific types
+        & returns a `ParsedChatCompletion` object, which is a subclass of the standard `ChatCompletion` class.
+
+        You can pass a pydantic model to this method and it will automatically convert the model
+        into a JSON schema, send it to the API and parse the response content back into the given model.
+
+        This method will also automatically parse `function` tool calls if:
+        - You use the `openai.pydantic_function_tool()` helper method
+        - You mark your tool schema with `"strict": True`
+
+        Example usage:
+        ```py
+        from pydantic import BaseModel
+        from openai import AsyncOpenAI
+
+
+        class Step(BaseModel):
+            explanation: str
+            output: str
+
+
+        class MathResponse(BaseModel):
+            steps: List[Step]
+            final_answer: str
+
+
+        client = AsyncOpenAI()
+        completion = await client.beta.chat.completions.parse(
+            model="gpt-4o-2024-08-06",
+            messages=[
+                {"role": "system", "content": "You are a helpful math tutor."},
+                {"role": "user", "content": "solve 8x + 31 = 2"},
+            ],
+            response_format=MathResponse,
+        )
+
+        message = completion.choices[0].message
+        if message.parsed:
+            print(message.parsed.steps)
+            print("answer: ", message.parsed.final_answer)
+        ```
+        """
+        _validate_input_tools(tools)
+
+        extra_headers = {
+            "X-Stainless-Helper-Method": "beta.chat.completions.parse",
+            **(extra_headers or {}),
+        }
+
+        def parser(raw_completion: ChatCompletion) -> ParsedChatCompletion[ResponseFormatT]:
+            return _parse_chat_completion(
+                response_format=response_format,
+                chat_completion=raw_completion,
+                input_tools=tools,
+            )
+
+        return await self._post(
+            "/chat/completions",
+            body=await async_maybe_transform(
+                {
+                    "messages": messages,
+                    "model": model,
+                    "frequency_penalty": frequency_penalty,
+                    "function_call": function_call,
+                    "functions": functions,
+                    "logit_bias": logit_bias,
+                    "logprobs": logprobs,
+                    "max_completion_tokens": max_completion_tokens,
+                    "max_tokens": max_tokens,
+                    "metadata": metadata,
+                    "n": n,
+                    "parallel_tool_calls": parallel_tool_calls,
+                    "presence_penalty": presence_penalty,
+                    "response_format": _type_to_response_format(response_format),
+                    "seed": seed,
+                    "service_tier": service_tier,
+                    "store": store,
+                    "stop": stop,
+                    "stream": False,
+                    "stream_options": stream_options,
+                    "temperature": temperature,
+                    "tool_choice": tool_choice,
+                    "tools": tools,
+                    "top_logprobs": top_logprobs,
+                    "top_p": top_p,
+                    "user": user,
+                },
+                completion_create_params.CompletionCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                post_parser=parser,
+            ),
+            # we turn the `ChatCompletion` instance into a `ParsedChatCompletion`
+            # in the `parser` function above
+            cast_to=cast(Type[ParsedChatCompletion[ResponseFormatT]], ChatCompletion),
+            stream=False,
+        )
+
+    def stream(
+        self,
+        *,
+        messages: Iterable[ChatCompletionMessageParam],
+        model: Union[str, ChatModel],
+        response_format: completion_create_params.ResponseFormat | type[ResponseFormatT] | NotGiven = NOT_GIVEN,
+        frequency_penalty: Optional[float] | NotGiven = NOT_GIVEN,
+        function_call: completion_create_params.FunctionCall | NotGiven = NOT_GIVEN,
+        functions: Iterable[completion_create_params.Function] | NotGiven = NOT_GIVEN,
+        logit_bias: Optional[Dict[str, int]] | NotGiven = NOT_GIVEN,
+        logprobs: Optional[bool] | NotGiven = NOT_GIVEN,
+        max_completion_tokens: Optional[int] | NotGiven = NOT_GIVEN,
+        max_tokens: Optional[int] | NotGiven = NOT_GIVEN,
+        metadata: Optional[Dict[str, str]] | NotGiven = NOT_GIVEN,
+        n: Optional[int] | NotGiven = NOT_GIVEN,
+        parallel_tool_calls: bool | NotGiven = NOT_GIVEN,
+        presence_penalty: Optional[float] | NotGiven = NOT_GIVEN,
+        seed: Optional[int] | NotGiven = NOT_GIVEN,
+        service_tier: Optional[Literal["auto", "default"]] | NotGiven = NOT_GIVEN,
+        stop: Union[Optional[str], List[str]] | NotGiven = NOT_GIVEN,
+        store: Optional[bool] | NotGiven = NOT_GIVEN,
+        stream_options: Optional[ChatCompletionStreamOptionsParam] | NotGiven = NOT_GIVEN,
+        temperature: Optional[float] | NotGiven = NOT_GIVEN,
+        tool_choice: ChatCompletionToolChoiceOptionParam | NotGiven = NOT_GIVEN,
+        tools: Iterable[ChatCompletionToolParam] | NotGiven = NOT_GIVEN,
+        top_logprobs: Optional[int] | NotGiven = NOT_GIVEN,
+        top_p: Optional[float] | NotGiven = NOT_GIVEN,
+        user: str | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> AsyncChatCompletionStreamManager[ResponseFormatT]:
+        """Wrapper over the `client.chat.completions.create(stream=True)` method that provides a more granular event API
+        and automatic accumulation of each delta.
+
+        This also supports all of the parsing utilities that `.parse()` does.
+
+        Unlike `.create(stream=True)`, the `.stream()` method requires usage within a context manager to prevent accidental leakage of the response:
+
+        ```py
+        async with client.beta.chat.completions.stream(
+            model="gpt-4o-2024-08-06",
+            messages=[...],
+        ) as stream:
+            async for event in stream:
+                if event.type == "content.delta":
+                    print(event.delta, flush=True, end="")
+        ```
+
+        When the context manager is entered, an `AsyncChatCompletionStream` instance is returned which, like `.create(stream=True)` is an async iterator. The full list of events that are yielded by the iterator are outlined in [these docs](https://github.com/openai/openai-python/blob/main/helpers.md#chat-completions-events).
+
+        When the context manager exits, the response will be closed, however the `stream` instance is still available outside
+        the context manager.
+        """
+        _validate_input_tools(tools)
+
+        extra_headers = {
+            "X-Stainless-Helper-Method": "beta.chat.completions.stream",
+            **(extra_headers or {}),
+        }
+
+        api_request = self._client.chat.completions.create(
+            messages=messages,
+            model=model,
+            stream=True,
+            response_format=_type_to_response_format(response_format),
+            frequency_penalty=frequency_penalty,
+            function_call=function_call,
+            functions=functions,
+            logit_bias=logit_bias,
+            logprobs=logprobs,
+            max_completion_tokens=max_completion_tokens,
+            max_tokens=max_tokens,
+            metadata=metadata,
+            n=n,
+            parallel_tool_calls=parallel_tool_calls,
+            presence_penalty=presence_penalty,
+            seed=seed,
+            service_tier=service_tier,
+            stop=stop,
+            store=store,
+            stream_options=stream_options,
+            temperature=temperature,
+            tool_choice=tool_choice,
+            tools=tools,
+            top_logprobs=top_logprobs,
+            top_p=top_p,
+            user=user,
+            extra_headers=extra_headers,
+            extra_query=extra_query,
+            extra_body=extra_body,
+            timeout=timeout,
+        )
+        return AsyncChatCompletionStreamManager(
+            api_request,
+            response_format=response_format,
+            input_tools=tools,
+        )
+
+
+class CompletionsWithRawResponse:
+    def __init__(self, completions: Completions) -> None:
+        self._completions = completions
+
+        self.parse = _legacy_response.to_raw_response_wrapper(
+            completions.parse,
+        )
+
+
+class AsyncCompletionsWithRawResponse:
+    def __init__(self, completions: AsyncCompletions) -> None:
+        self._completions = completions
+
+        self.parse = _legacy_response.async_to_raw_response_wrapper(
+            completions.parse,
+        )
+
+
+class CompletionsWithStreamingResponse:
+    def __init__(self, completions: Completions) -> None:
+        self._completions = completions
+
+        self.parse = to_streamed_response_wrapper(
+            completions.parse,
+        )
+
+
+class AsyncCompletionsWithStreamingResponse:
+    def __init__(self, completions: AsyncCompletions) -> None:
+        self._completions = completions
+
+        self.parse = async_to_streamed_response_wrapper(
+            completions.parse,
+        )

backend/venv/lib/python3.12/site-packages/openai/resources/beta/threads/__init__.py

@@ -0,0 +1,47 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from .runs import (
+    Runs,
+    AsyncRuns,
+    RunsWithRawResponse,
+    AsyncRunsWithRawResponse,
+    RunsWithStreamingResponse,
+    AsyncRunsWithStreamingResponse,
+)
+from .threads import (
+    Threads,
+    AsyncThreads,
+    ThreadsWithRawResponse,
+    AsyncThreadsWithRawResponse,
+    ThreadsWithStreamingResponse,
+    AsyncThreadsWithStreamingResponse,
+)
+from .messages import (
+    Messages,
+    AsyncMessages,
+    MessagesWithRawResponse,
+    AsyncMessagesWithRawResponse,
+    MessagesWithStreamingResponse,
+    AsyncMessagesWithStreamingResponse,
+)
+
+__all__ = [
+    "Runs",
+    "AsyncRuns",
+    "RunsWithRawResponse",
+    "AsyncRunsWithRawResponse",
+    "RunsWithStreamingResponse",
+    "AsyncRunsWithStreamingResponse",
+    "Messages",
+    "AsyncMessages",
+    "MessagesWithRawResponse",
+    "AsyncMessagesWithRawResponse",
+    "MessagesWithStreamingResponse",
+    "AsyncMessagesWithStreamingResponse",
+    "Threads",
+    "AsyncThreads",
+    "ThreadsWithRawResponse",
+    "AsyncThreadsWithRawResponse",
+    "ThreadsWithStreamingResponse",
+    "AsyncThreadsWithStreamingResponse",
+]

backend/venv/lib/python3.12/site-packages/openai/resources/beta/threads/messages.py

@@ -0,0 +1,661 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Union, Iterable, Optional
+from typing_extensions import Literal
+
+import httpx
+
+from .... import _legacy_response
+from ...._types import NOT_GIVEN, Body, Query, Headers, NotGiven
+from ...._utils import (
+    maybe_transform,
+    async_maybe_transform,
+)
+from ...._compat import cached_property
+from ...._resource import SyncAPIResource, AsyncAPIResource
+from ...._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
+from ....pagination import SyncCursorPage, AsyncCursorPage
+from ...._base_client import (
+    AsyncPaginator,
+    make_request_options,
+)
+from ....types.beta.threads import message_list_params, message_create_params, message_update_params
+from ....types.beta.threads.message import Message
+from ....types.beta.threads.message_deleted import MessageDeleted
+from ....types.beta.threads.message_content_part_param import MessageContentPartParam
+
+__all__ = ["Messages", "AsyncMessages"]
+
+
+class Messages(SyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> MessagesWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return MessagesWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> MessagesWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return MessagesWithStreamingResponse(self)
+
+    def create(
+        self,
+        thread_id: str,
+        *,
+        content: Union[str, Iterable[MessageContentPartParam]],
+        role: Literal["user", "assistant"],
+        attachments: Optional[Iterable[message_create_params.Attachment]] | NotGiven = NOT_GIVEN,
+        metadata: Optional[object] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Message:
+        """
+        Create a message.
+
+        Args:
+          content: The text contents of the message.
+
+          role:
+              The role of the entity that is creating the message. Allowed values include:
+
+              - `user`: Indicates the message is sent by an actual user and should be used in
+                most cases to represent user-generated messages.
+              - `assistant`: Indicates the message is generated by the assistant. Use this
+                value to insert messages from the assistant into the conversation.
+
+          attachments: A list of files attached to the message, and the tools they should be added to.
+
+          metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
+              for storing additional information about the object in a structured format. Keys
+              can be a maximum of 64 characters long and values can be a maximum of 512
+              characters long.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not thread_id:
+            raise ValueError(f"Expected a non-empty value for `thread_id` but received {thread_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._post(
+            f"/threads/{thread_id}/messages",
+            body=maybe_transform(
+                {
+                    "content": content,
+                    "role": role,
+                    "attachments": attachments,
+                    "metadata": metadata,
+                },
+                message_create_params.MessageCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=Message,
+        )
+
+    def retrieve(
+        self,
+        message_id: str,
+        *,
+        thread_id: str,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Message:
+        """
+        Retrieve a message.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not thread_id:
+            raise ValueError(f"Expected a non-empty value for `thread_id` but received {thread_id!r}")
+        if not message_id:
+            raise ValueError(f"Expected a non-empty value for `message_id` but received {message_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._get(
+            f"/threads/{thread_id}/messages/{message_id}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=Message,
+        )
+
+    def update(
+        self,
+        message_id: str,
+        *,
+        thread_id: str,
+        metadata: Optional[object] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Message:
+        """
+        Modifies a message.
+
+        Args:
+          metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
+              for storing additional information about the object in a structured format. Keys
+              can be a maximum of 64 characters long and values can be a maximum of 512
+              characters long.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not thread_id:
+            raise ValueError(f"Expected a non-empty value for `thread_id` but received {thread_id!r}")
+        if not message_id:
+            raise ValueError(f"Expected a non-empty value for `message_id` but received {message_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._post(
+            f"/threads/{thread_id}/messages/{message_id}",
+            body=maybe_transform({"metadata": metadata}, message_update_params.MessageUpdateParams),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=Message,
+        )
+
+    def list(
+        self,
+        thread_id: str,
+        *,
+        after: str | NotGiven = NOT_GIVEN,
+        before: str | NotGiven = NOT_GIVEN,
+        limit: int | NotGiven = NOT_GIVEN,
+        order: Literal["asc", "desc"] | NotGiven = NOT_GIVEN,
+        run_id: str | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> SyncCursorPage[Message]:
+        """
+        Returns a list of messages for a given thread.
+
+        Args:
+          after: A cursor for use in pagination. `after` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include after=obj_foo in order to
+              fetch the next page of the list.
+
+          before: A cursor for use in pagination. `before` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include before=obj_foo in order to
+              fetch the previous page of the list.
+
+          limit: A limit on the number of objects to be returned. Limit can range between 1 and
+              100, and the default is 20.
+
+          order: Sort order by the `created_at` timestamp of the objects. `asc` for ascending
+              order and `desc` for descending order.
+
+          run_id: Filter messages by the run ID that generated them.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not thread_id:
+            raise ValueError(f"Expected a non-empty value for `thread_id` but received {thread_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._get_api_list(
+            f"/threads/{thread_id}/messages",
+            page=SyncCursorPage[Message],
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform(
+                    {
+                        "after": after,
+                        "before": before,
+                        "limit": limit,
+                        "order": order,
+                        "run_id": run_id,
+                    },
+                    message_list_params.MessageListParams,
+                ),
+            ),
+            model=Message,
+        )
+
+    def delete(
+        self,
+        message_id: str,
+        *,
+        thread_id: str,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> MessageDeleted:
+        """
+        Deletes a message.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not thread_id:
+            raise ValueError(f"Expected a non-empty value for `thread_id` but received {thread_id!r}")
+        if not message_id:
+            raise ValueError(f"Expected a non-empty value for `message_id` but received {message_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._delete(
+            f"/threads/{thread_id}/messages/{message_id}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=MessageDeleted,
+        )
+
+
+class AsyncMessages(AsyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> AsyncMessagesWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncMessagesWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncMessagesWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncMessagesWithStreamingResponse(self)
+
+    async def create(
+        self,
+        thread_id: str,
+        *,
+        content: Union[str, Iterable[MessageContentPartParam]],
+        role: Literal["user", "assistant"],
+        attachments: Optional[Iterable[message_create_params.Attachment]] | NotGiven = NOT_GIVEN,
+        metadata: Optional[object] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Message:
+        """
+        Create a message.
+
+        Args:
+          content: The text contents of the message.
+
+          role:
+              The role of the entity that is creating the message. Allowed values include:
+
+              - `user`: Indicates the message is sent by an actual user and should be used in
+                most cases to represent user-generated messages.
+              - `assistant`: Indicates the message is generated by the assistant. Use this
+                value to insert messages from the assistant into the conversation.
+
+          attachments: A list of files attached to the message, and the tools they should be added to.
+
+          metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
+              for storing additional information about the object in a structured format. Keys
+              can be a maximum of 64 characters long and values can be a maximum of 512
+              characters long.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not thread_id:
+            raise ValueError(f"Expected a non-empty value for `thread_id` but received {thread_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return await self._post(
+            f"/threads/{thread_id}/messages",
+            body=await async_maybe_transform(
+                {
+                    "content": content,
+                    "role": role,
+                    "attachments": attachments,
+                    "metadata": metadata,
+                },
+                message_create_params.MessageCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=Message,
+        )
+
+    async def retrieve(
+        self,
+        message_id: str,
+        *,
+        thread_id: str,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Message:
+        """
+        Retrieve a message.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not thread_id:
+            raise ValueError(f"Expected a non-empty value for `thread_id` but received {thread_id!r}")
+        if not message_id:
+            raise ValueError(f"Expected a non-empty value for `message_id` but received {message_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return await self._get(
+            f"/threads/{thread_id}/messages/{message_id}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=Message,
+        )
+
+    async def update(
+        self,
+        message_id: str,
+        *,
+        thread_id: str,
+        metadata: Optional[object] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Message:
+        """
+        Modifies a message.
+
+        Args:
+          metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
+              for storing additional information about the object in a structured format. Keys
+              can be a maximum of 64 characters long and values can be a maximum of 512
+              characters long.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not thread_id:
+            raise ValueError(f"Expected a non-empty value for `thread_id` but received {thread_id!r}")
+        if not message_id:
+            raise ValueError(f"Expected a non-empty value for `message_id` but received {message_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return await self._post(
+            f"/threads/{thread_id}/messages/{message_id}",
+            body=await async_maybe_transform({"metadata": metadata}, message_update_params.MessageUpdateParams),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=Message,
+        )
+
+    def list(
+        self,
+        thread_id: str,
+        *,
+        after: str | NotGiven = NOT_GIVEN,
+        before: str | NotGiven = NOT_GIVEN,
+        limit: int | NotGiven = NOT_GIVEN,
+        order: Literal["asc", "desc"] | NotGiven = NOT_GIVEN,
+        run_id: str | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> AsyncPaginator[Message, AsyncCursorPage[Message]]:
+        """
+        Returns a list of messages for a given thread.
+
+        Args:
+          after: A cursor for use in pagination. `after` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include after=obj_foo in order to
+              fetch the next page of the list.
+
+          before: A cursor for use in pagination. `before` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include before=obj_foo in order to
+              fetch the previous page of the list.
+
+          limit: A limit on the number of objects to be returned. Limit can range between 1 and
+              100, and the default is 20.
+
+          order: Sort order by the `created_at` timestamp of the objects. `asc` for ascending
+              order and `desc` for descending order.
+
+          run_id: Filter messages by the run ID that generated them.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not thread_id:
+            raise ValueError(f"Expected a non-empty value for `thread_id` but received {thread_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._get_api_list(
+            f"/threads/{thread_id}/messages",
+            page=AsyncCursorPage[Message],
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform(
+                    {
+                        "after": after,
+                        "before": before,
+                        "limit": limit,
+                        "order": order,
+                        "run_id": run_id,
+                    },
+                    message_list_params.MessageListParams,
+                ),
+            ),
+            model=Message,
+        )
+
+    async def delete(
+        self,
+        message_id: str,
+        *,
+        thread_id: str,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> MessageDeleted:
+        """
+        Deletes a message.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not thread_id:
+            raise ValueError(f"Expected a non-empty value for `thread_id` but received {thread_id!r}")
+        if not message_id:
+            raise ValueError(f"Expected a non-empty value for `message_id` but received {message_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return await self._delete(
+            f"/threads/{thread_id}/messages/{message_id}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=MessageDeleted,
+        )
+
+
+class MessagesWithRawResponse:
+    def __init__(self, messages: Messages) -> None:
+        self._messages = messages
+
+        self.create = _legacy_response.to_raw_response_wrapper(
+            messages.create,
+        )
+        self.retrieve = _legacy_response.to_raw_response_wrapper(
+            messages.retrieve,
+        )
+        self.update = _legacy_response.to_raw_response_wrapper(
+            messages.update,
+        )
+        self.list = _legacy_response.to_raw_response_wrapper(
+            messages.list,
+        )
+        self.delete = _legacy_response.to_raw_response_wrapper(
+            messages.delete,
+        )
+
+
+class AsyncMessagesWithRawResponse:
+    def __init__(self, messages: AsyncMessages) -> None:
+        self._messages = messages
+
+        self.create = _legacy_response.async_to_raw_response_wrapper(
+            messages.create,
+        )
+        self.retrieve = _legacy_response.async_to_raw_response_wrapper(
+            messages.retrieve,
+        )
+        self.update = _legacy_response.async_to_raw_response_wrapper(
+            messages.update,
+        )
+        self.list = _legacy_response.async_to_raw_response_wrapper(
+            messages.list,
+        )
+        self.delete = _legacy_response.async_to_raw_response_wrapper(
+            messages.delete,
+        )
+
+
+class MessagesWithStreamingResponse:
+    def __init__(self, messages: Messages) -> None:
+        self._messages = messages
+
+        self.create = to_streamed_response_wrapper(
+            messages.create,
+        )
+        self.retrieve = to_streamed_response_wrapper(
+            messages.retrieve,
+        )
+        self.update = to_streamed_response_wrapper(
+            messages.update,
+        )
+        self.list = to_streamed_response_wrapper(
+            messages.list,
+        )
+        self.delete = to_streamed_response_wrapper(
+            messages.delete,
+        )
+
+
+class AsyncMessagesWithStreamingResponse:
+    def __init__(self, messages: AsyncMessages) -> None:
+        self._messages = messages
+
+        self.create = async_to_streamed_response_wrapper(
+            messages.create,
+        )
+        self.retrieve = async_to_streamed_response_wrapper(
+            messages.retrieve,
+        )
+        self.update = async_to_streamed_response_wrapper(
+            messages.update,
+        )
+        self.list = async_to_streamed_response_wrapper(
+            messages.list,
+        )
+        self.delete = async_to_streamed_response_wrapper(
+            messages.delete,
+        )

backend/venv/lib/python3.12/site-packages/openai/resources/beta/threads/runs/__init__.py

@@ -0,0 +1,33 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from .runs import (
+    Runs,
+    AsyncRuns,
+    RunsWithRawResponse,
+    AsyncRunsWithRawResponse,
+    RunsWithStreamingResponse,
+    AsyncRunsWithStreamingResponse,
+)
+from .steps import (
+    Steps,
+    AsyncSteps,
+    StepsWithRawResponse,
+    AsyncStepsWithRawResponse,
+    StepsWithStreamingResponse,
+    AsyncStepsWithStreamingResponse,
+)
+
+__all__ = [
+    "Steps",
+    "AsyncSteps",
+    "StepsWithRawResponse",
+    "AsyncStepsWithRawResponse",
+    "StepsWithStreamingResponse",
+    "AsyncStepsWithStreamingResponse",
+    "Runs",
+    "AsyncRuns",
+    "RunsWithRawResponse",
+    "AsyncRunsWithRawResponse",
+    "RunsWithStreamingResponse",
+    "AsyncRunsWithStreamingResponse",
+]

backend/venv/lib/python3.12/site-packages/openai/resources/beta/threads/runs/steps.py

@@ -0,0 +1,381 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import List
+from typing_extensions import Literal
+
+import httpx
+
+from ..... import _legacy_response
+from ....._types import NOT_GIVEN, Body, Query, Headers, NotGiven
+from ....._utils import (
+    maybe_transform,
+    async_maybe_transform,
+)
+from ....._compat import cached_property
+from ....._resource import SyncAPIResource, AsyncAPIResource
+from ....._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
+from .....pagination import SyncCursorPage, AsyncCursorPage
+from ....._base_client import AsyncPaginator, make_request_options
+from .....types.beta.threads.runs import step_list_params, step_retrieve_params
+from .....types.beta.threads.runs.run_step import RunStep
+from .....types.beta.threads.runs.run_step_include import RunStepInclude
+
+__all__ = ["Steps", "AsyncSteps"]
+
+
+class Steps(SyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> StepsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return StepsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> StepsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return StepsWithStreamingResponse(self)
+
+    def retrieve(
+        self,
+        step_id: str,
+        *,
+        thread_id: str,
+        run_id: str,
+        include: List[RunStepInclude] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> RunStep:
+        """
+        Retrieves a run step.
+
+        Args:
+          include: A list of additional fields to include in the response. Currently the only
+              supported value is `step_details.tool_calls[*].file_search.results[*].content`
+              to fetch the file search result content.
+
+              See the
+              [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search/customizing-file-search-settings)
+              for more information.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not thread_id:
+            raise ValueError(f"Expected a non-empty value for `thread_id` but received {thread_id!r}")
+        if not run_id:
+            raise ValueError(f"Expected a non-empty value for `run_id` but received {run_id!r}")
+        if not step_id:
+            raise ValueError(f"Expected a non-empty value for `step_id` but received {step_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._get(
+            f"/threads/{thread_id}/runs/{run_id}/steps/{step_id}",
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform({"include": include}, step_retrieve_params.StepRetrieveParams),
+            ),
+            cast_to=RunStep,
+        )
+
+    def list(
+        self,
+        run_id: str,
+        *,
+        thread_id: str,
+        after: str | NotGiven = NOT_GIVEN,
+        before: str | NotGiven = NOT_GIVEN,
+        include: List[RunStepInclude] | NotGiven = NOT_GIVEN,
+        limit: int | NotGiven = NOT_GIVEN,
+        order: Literal["asc", "desc"] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> SyncCursorPage[RunStep]:
+        """
+        Returns a list of run steps belonging to a run.
+
+        Args:
+          after: A cursor for use in pagination. `after` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include after=obj_foo in order to
+              fetch the next page of the list.
+
+          before: A cursor for use in pagination. `before` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include before=obj_foo in order to
+              fetch the previous page of the list.
+
+          include: A list of additional fields to include in the response. Currently the only
+              supported value is `step_details.tool_calls[*].file_search.results[*].content`
+              to fetch the file search result content.
+
+              See the
+              [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search/customizing-file-search-settings)
+              for more information.
+
+          limit: A limit on the number of objects to be returned. Limit can range between 1 and
+              100, and the default is 20.
+
+          order: Sort order by the `created_at` timestamp of the objects. `asc` for ascending
+              order and `desc` for descending order.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not thread_id:
+            raise ValueError(f"Expected a non-empty value for `thread_id` but received {thread_id!r}")
+        if not run_id:
+            raise ValueError(f"Expected a non-empty value for `run_id` but received {run_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._get_api_list(
+            f"/threads/{thread_id}/runs/{run_id}/steps",
+            page=SyncCursorPage[RunStep],
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform(
+                    {
+                        "after": after,
+                        "before": before,
+                        "include": include,
+                        "limit": limit,
+                        "order": order,
+                    },
+                    step_list_params.StepListParams,
+                ),
+            ),
+            model=RunStep,
+        )
+
+
+class AsyncSteps(AsyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> AsyncStepsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncStepsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncStepsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncStepsWithStreamingResponse(self)
+
+    async def retrieve(
+        self,
+        step_id: str,
+        *,
+        thread_id: str,
+        run_id: str,
+        include: List[RunStepInclude] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> RunStep:
+        """
+        Retrieves a run step.
+
+        Args:
+          include: A list of additional fields to include in the response. Currently the only
+              supported value is `step_details.tool_calls[*].file_search.results[*].content`
+              to fetch the file search result content.
+
+              See the
+              [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search/customizing-file-search-settings)
+              for more information.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not thread_id:
+            raise ValueError(f"Expected a non-empty value for `thread_id` but received {thread_id!r}")
+        if not run_id:
+            raise ValueError(f"Expected a non-empty value for `run_id` but received {run_id!r}")
+        if not step_id:
+            raise ValueError(f"Expected a non-empty value for `step_id` but received {step_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return await self._get(
+            f"/threads/{thread_id}/runs/{run_id}/steps/{step_id}",
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=await async_maybe_transform({"include": include}, step_retrieve_params.StepRetrieveParams),
+            ),
+            cast_to=RunStep,
+        )
+
+    def list(
+        self,
+        run_id: str,
+        *,
+        thread_id: str,
+        after: str | NotGiven = NOT_GIVEN,
+        before: str | NotGiven = NOT_GIVEN,
+        include: List[RunStepInclude] | NotGiven = NOT_GIVEN,
+        limit: int | NotGiven = NOT_GIVEN,
+        order: Literal["asc", "desc"] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> AsyncPaginator[RunStep, AsyncCursorPage[RunStep]]:
+        """
+        Returns a list of run steps belonging to a run.
+
+        Args:
+          after: A cursor for use in pagination. `after` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include after=obj_foo in order to
+              fetch the next page of the list.
+
+          before: A cursor for use in pagination. `before` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include before=obj_foo in order to
+              fetch the previous page of the list.
+
+          include: A list of additional fields to include in the response. Currently the only
+              supported value is `step_details.tool_calls[*].file_search.results[*].content`
+              to fetch the file search result content.
+
+              See the
+              [file search tool documentation](https://platform.openai.com/docs/assistants/tools/file-search/customizing-file-search-settings)
+              for more information.
+
+          limit: A limit on the number of objects to be returned. Limit can range between 1 and
+              100, and the default is 20.
+
+          order: Sort order by the `created_at` timestamp of the objects. `asc` for ascending
+              order and `desc` for descending order.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not thread_id:
+            raise ValueError(f"Expected a non-empty value for `thread_id` but received {thread_id!r}")
+        if not run_id:
+            raise ValueError(f"Expected a non-empty value for `run_id` but received {run_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._get_api_list(
+            f"/threads/{thread_id}/runs/{run_id}/steps",
+            page=AsyncCursorPage[RunStep],
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform(
+                    {
+                        "after": after,
+                        "before": before,
+                        "include": include,
+                        "limit": limit,
+                        "order": order,
+                    },
+                    step_list_params.StepListParams,
+                ),
+            ),
+            model=RunStep,
+        )
+
+
+class StepsWithRawResponse:
+    def __init__(self, steps: Steps) -> None:
+        self._steps = steps
+
+        self.retrieve = _legacy_response.to_raw_response_wrapper(
+            steps.retrieve,
+        )
+        self.list = _legacy_response.to_raw_response_wrapper(
+            steps.list,
+        )
+
+
+class AsyncStepsWithRawResponse:
+    def __init__(self, steps: AsyncSteps) -> None:
+        self._steps = steps
+
+        self.retrieve = _legacy_response.async_to_raw_response_wrapper(
+            steps.retrieve,
+        )
+        self.list = _legacy_response.async_to_raw_response_wrapper(
+            steps.list,
+        )
+
+
+class StepsWithStreamingResponse:
+    def __init__(self, steps: Steps) -> None:
+        self._steps = steps
+
+        self.retrieve = to_streamed_response_wrapper(
+            steps.retrieve,
+        )
+        self.list = to_streamed_response_wrapper(
+            steps.list,
+        )
+
+
+class AsyncStepsWithStreamingResponse:
+    def __init__(self, steps: AsyncSteps) -> None:
+        self._steps = steps
+
+        self.retrieve = async_to_streamed_response_wrapper(
+            steps.retrieve,
+        )
+        self.list = async_to_streamed_response_wrapper(
+            steps.list,
+        )

backend/venv/lib/python3.12/site-packages/openai/resources/beta/vector_stores/__init__.py

@@ -0,0 +1,47 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from .files import (
+    Files,
+    AsyncFiles,
+    FilesWithRawResponse,
+    AsyncFilesWithRawResponse,
+    FilesWithStreamingResponse,
+    AsyncFilesWithStreamingResponse,
+)
+from .file_batches import (
+    FileBatches,
+    AsyncFileBatches,
+    FileBatchesWithRawResponse,
+    AsyncFileBatchesWithRawResponse,
+    FileBatchesWithStreamingResponse,
+    AsyncFileBatchesWithStreamingResponse,
+)
+from .vector_stores import (
+    VectorStores,
+    AsyncVectorStores,
+    VectorStoresWithRawResponse,
+    AsyncVectorStoresWithRawResponse,
+    VectorStoresWithStreamingResponse,
+    AsyncVectorStoresWithStreamingResponse,
+)
+
+__all__ = [
+    "Files",
+    "AsyncFiles",
+    "FilesWithRawResponse",
+    "AsyncFilesWithRawResponse",
+    "FilesWithStreamingResponse",
+    "AsyncFilesWithStreamingResponse",
+    "FileBatches",
+    "AsyncFileBatches",
+    "FileBatchesWithRawResponse",
+    "AsyncFileBatchesWithRawResponse",
+    "FileBatchesWithStreamingResponse",
+    "AsyncFileBatchesWithStreamingResponse",
+    "VectorStores",
+    "AsyncVectorStores",
+    "VectorStoresWithRawResponse",
+    "AsyncVectorStoresWithRawResponse",
+    "VectorStoresWithStreamingResponse",
+    "AsyncVectorStoresWithStreamingResponse",
+]

backend/venv/lib/python3.12/site-packages/openai/resources/beta/vector_stores/file_batches.py

@@ -0,0 +1,785 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+import asyncio
+from typing import List, Iterable
+from typing_extensions import Literal
+from concurrent.futures import Future, ThreadPoolExecutor, as_completed
+
+import httpx
+import sniffio
+
+from .... import _legacy_response
+from ....types import FileObject
+from ...._types import NOT_GIVEN, Body, Query, Headers, NotGiven, FileTypes
+from ...._utils import (
+    is_given,
+    maybe_transform,
+    async_maybe_transform,
+)
+from ...._compat import cached_property
+from ...._resource import SyncAPIResource, AsyncAPIResource
+from ...._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
+from ....pagination import SyncCursorPage, AsyncCursorPage
+from ....types.beta import FileChunkingStrategyParam
+from ...._base_client import AsyncPaginator, make_request_options
+from ....types.beta.vector_stores import file_batch_create_params, file_batch_list_files_params
+from ....types.beta.file_chunking_strategy_param import FileChunkingStrategyParam
+from ....types.beta.vector_stores.vector_store_file import VectorStoreFile
+from ....types.beta.vector_stores.vector_store_file_batch import VectorStoreFileBatch
+
+__all__ = ["FileBatches", "AsyncFileBatches"]
+
+
+class FileBatches(SyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> FileBatchesWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return FileBatchesWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> FileBatchesWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return FileBatchesWithStreamingResponse(self)
+
+    def create(
+        self,
+        vector_store_id: str,
+        *,
+        file_ids: List[str],
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreFileBatch:
+        """
+        Create a vector store file batch.
+
+        Args:
+          file_ids: A list of [File](https://platform.openai.com/docs/api-reference/files) IDs that
+              the vector store should use. Useful for tools like `file_search` that can access
+              files.
+
+          chunking_strategy: The chunking strategy used to chunk the file(s). If not set, will use the `auto`
+              strategy. Only applicable if `file_ids` is non-empty.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not vector_store_id:
+            raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._post(
+            f"/vector_stores/{vector_store_id}/file_batches",
+            body=maybe_transform(
+                {
+                    "file_ids": file_ids,
+                    "chunking_strategy": chunking_strategy,
+                },
+                file_batch_create_params.FileBatchCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=VectorStoreFileBatch,
+        )
+
+    def retrieve(
+        self,
+        batch_id: str,
+        *,
+        vector_store_id: str,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreFileBatch:
+        """
+        Retrieves a vector store file batch.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not vector_store_id:
+            raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}")
+        if not batch_id:
+            raise ValueError(f"Expected a non-empty value for `batch_id` but received {batch_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._get(
+            f"/vector_stores/{vector_store_id}/file_batches/{batch_id}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=VectorStoreFileBatch,
+        )
+
+    def cancel(
+        self,
+        batch_id: str,
+        *,
+        vector_store_id: str,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreFileBatch:
+        """Cancel a vector store file batch.
+
+        This attempts to cancel the processing of
+        files in this batch as soon as possible.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not vector_store_id:
+            raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}")
+        if not batch_id:
+            raise ValueError(f"Expected a non-empty value for `batch_id` but received {batch_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._post(
+            f"/vector_stores/{vector_store_id}/file_batches/{batch_id}/cancel",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=VectorStoreFileBatch,
+        )
+
+    def create_and_poll(
+        self,
+        vector_store_id: str,
+        *,
+        file_ids: List[str],
+        poll_interval_ms: int | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreFileBatch:
+        """Create a vector store batch and poll until all files have been processed."""
+        batch = self.create(
+            vector_store_id=vector_store_id,
+            file_ids=file_ids,
+            chunking_strategy=chunking_strategy,
+        )
+        # TODO: don't poll unless necessary??
+        return self.poll(
+            batch.id,
+            vector_store_id=vector_store_id,
+            poll_interval_ms=poll_interval_ms,
+        )
+
+    def list_files(
+        self,
+        batch_id: str,
+        *,
+        vector_store_id: str,
+        after: str | NotGiven = NOT_GIVEN,
+        before: str | NotGiven = NOT_GIVEN,
+        filter: Literal["in_progress", "completed", "failed", "cancelled"] | NotGiven = NOT_GIVEN,
+        limit: int | NotGiven = NOT_GIVEN,
+        order: Literal["asc", "desc"] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> SyncCursorPage[VectorStoreFile]:
+        """
+        Returns a list of vector store files in a batch.
+
+        Args:
+          after: A cursor for use in pagination. `after` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include after=obj_foo in order to
+              fetch the next page of the list.
+
+          before: A cursor for use in pagination. `before` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include before=obj_foo in order to
+              fetch the previous page of the list.
+
+          filter: Filter by file status. One of `in_progress`, `completed`, `failed`, `cancelled`.
+
+          limit: A limit on the number of objects to be returned. Limit can range between 1 and
+              100, and the default is 20.
+
+          order: Sort order by the `created_at` timestamp of the objects. `asc` for ascending
+              order and `desc` for descending order.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not vector_store_id:
+            raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}")
+        if not batch_id:
+            raise ValueError(f"Expected a non-empty value for `batch_id` but received {batch_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._get_api_list(
+            f"/vector_stores/{vector_store_id}/file_batches/{batch_id}/files",
+            page=SyncCursorPage[VectorStoreFile],
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform(
+                    {
+                        "after": after,
+                        "before": before,
+                        "filter": filter,
+                        "limit": limit,
+                        "order": order,
+                    },
+                    file_batch_list_files_params.FileBatchListFilesParams,
+                ),
+            ),
+            model=VectorStoreFile,
+        )
+
+    def poll(
+        self,
+        batch_id: str,
+        *,
+        vector_store_id: str,
+        poll_interval_ms: int | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreFileBatch:
+        """Wait for the given file batch to be processed.
+
+        Note: this will return even if one of the files failed to process, you need to
+        check batch.file_counts.failed_count to handle this case.
+        """
+        headers: dict[str, str] = {"X-Stainless-Poll-Helper": "true"}
+        if is_given(poll_interval_ms):
+            headers["X-Stainless-Custom-Poll-Interval"] = str(poll_interval_ms)
+
+        while True:
+            response = self.with_raw_response.retrieve(
+                batch_id,
+                vector_store_id=vector_store_id,
+                extra_headers=headers,
+            )
+
+            batch = response.parse()
+            if batch.file_counts.in_progress > 0:
+                if not is_given(poll_interval_ms):
+                    from_header = response.headers.get("openai-poll-after-ms")
+                    if from_header is not None:
+                        poll_interval_ms = int(from_header)
+                    else:
+                        poll_interval_ms = 1000
+
+                self._sleep(poll_interval_ms / 1000)
+                continue
+
+            return batch
+
+    def upload_and_poll(
+        self,
+        vector_store_id: str,
+        *,
+        files: Iterable[FileTypes],
+        max_concurrency: int = 5,
+        file_ids: List[str] = [],
+        poll_interval_ms: int | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreFileBatch:
+        """Uploads the given files concurrently and then creates a vector store file batch.
+
+        If you've already uploaded certain files that you want to include in this batch
+        then you can pass their IDs through the `file_ids` argument.
+
+        By default, if any file upload fails then an exception will be eagerly raised.
+
+        The number of concurrency uploads is configurable using the `max_concurrency`
+        parameter.
+
+        Note: this method only supports `asyncio` or `trio` as the backing async
+        runtime.
+        """
+        results: list[FileObject] = []
+
+        with ThreadPoolExecutor(max_workers=max_concurrency) as executor:
+            futures: list[Future[FileObject]] = [
+                executor.submit(
+                    self._client.files.create,
+                    file=file,
+                    purpose="assistants",
+                )
+                for file in files
+            ]
+
+        for future in as_completed(futures):
+            exc = future.exception()
+            if exc:
+                raise exc
+
+            results.append(future.result())
+
+        batch = self.create_and_poll(
+            vector_store_id=vector_store_id,
+            file_ids=[*file_ids, *(f.id for f in results)],
+            poll_interval_ms=poll_interval_ms,
+            chunking_strategy=chunking_strategy,
+        )
+        return batch
+
+
+class AsyncFileBatches(AsyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> AsyncFileBatchesWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncFileBatchesWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncFileBatchesWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncFileBatchesWithStreamingResponse(self)
+
+    async def create(
+        self,
+        vector_store_id: str,
+        *,
+        file_ids: List[str],
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreFileBatch:
+        """
+        Create a vector store file batch.
+
+        Args:
+          file_ids: A list of [File](https://platform.openai.com/docs/api-reference/files) IDs that
+              the vector store should use. Useful for tools like `file_search` that can access
+              files.
+
+          chunking_strategy: The chunking strategy used to chunk the file(s). If not set, will use the `auto`
+              strategy. Only applicable if `file_ids` is non-empty.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not vector_store_id:
+            raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return await self._post(
+            f"/vector_stores/{vector_store_id}/file_batches",
+            body=await async_maybe_transform(
+                {
+                    "file_ids": file_ids,
+                    "chunking_strategy": chunking_strategy,
+                },
+                file_batch_create_params.FileBatchCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=VectorStoreFileBatch,
+        )
+
+    async def retrieve(
+        self,
+        batch_id: str,
+        *,
+        vector_store_id: str,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreFileBatch:
+        """
+        Retrieves a vector store file batch.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not vector_store_id:
+            raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}")
+        if not batch_id:
+            raise ValueError(f"Expected a non-empty value for `batch_id` but received {batch_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return await self._get(
+            f"/vector_stores/{vector_store_id}/file_batches/{batch_id}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=VectorStoreFileBatch,
+        )
+
+    async def cancel(
+        self,
+        batch_id: str,
+        *,
+        vector_store_id: str,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreFileBatch:
+        """Cancel a vector store file batch.
+
+        This attempts to cancel the processing of
+        files in this batch as soon as possible.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not vector_store_id:
+            raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}")
+        if not batch_id:
+            raise ValueError(f"Expected a non-empty value for `batch_id` but received {batch_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return await self._post(
+            f"/vector_stores/{vector_store_id}/file_batches/{batch_id}/cancel",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=VectorStoreFileBatch,
+        )
+
+    async def create_and_poll(
+        self,
+        vector_store_id: str,
+        *,
+        file_ids: List[str],
+        poll_interval_ms: int | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreFileBatch:
+        """Create a vector store batch and poll until all files have been processed."""
+        batch = await self.create(
+            vector_store_id=vector_store_id,
+            file_ids=file_ids,
+            chunking_strategy=chunking_strategy,
+        )
+        # TODO: don't poll unless necessary??
+        return await self.poll(
+            batch.id,
+            vector_store_id=vector_store_id,
+            poll_interval_ms=poll_interval_ms,
+        )
+
+    def list_files(
+        self,
+        batch_id: str,
+        *,
+        vector_store_id: str,
+        after: str | NotGiven = NOT_GIVEN,
+        before: str | NotGiven = NOT_GIVEN,
+        filter: Literal["in_progress", "completed", "failed", "cancelled"] | NotGiven = NOT_GIVEN,
+        limit: int | NotGiven = NOT_GIVEN,
+        order: Literal["asc", "desc"] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> AsyncPaginator[VectorStoreFile, AsyncCursorPage[VectorStoreFile]]:
+        """
+        Returns a list of vector store files in a batch.
+
+        Args:
+          after: A cursor for use in pagination. `after` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include after=obj_foo in order to
+              fetch the next page of the list.
+
+          before: A cursor for use in pagination. `before` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include before=obj_foo in order to
+              fetch the previous page of the list.
+
+          filter: Filter by file status. One of `in_progress`, `completed`, `failed`, `cancelled`.
+
+          limit: A limit on the number of objects to be returned. Limit can range between 1 and
+              100, and the default is 20.
+
+          order: Sort order by the `created_at` timestamp of the objects. `asc` for ascending
+              order and `desc` for descending order.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not vector_store_id:
+            raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}")
+        if not batch_id:
+            raise ValueError(f"Expected a non-empty value for `batch_id` but received {batch_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._get_api_list(
+            f"/vector_stores/{vector_store_id}/file_batches/{batch_id}/files",
+            page=AsyncCursorPage[VectorStoreFile],
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform(
+                    {
+                        "after": after,
+                        "before": before,
+                        "filter": filter,
+                        "limit": limit,
+                        "order": order,
+                    },
+                    file_batch_list_files_params.FileBatchListFilesParams,
+                ),
+            ),
+            model=VectorStoreFile,
+        )
+
+    async def poll(
+        self,
+        batch_id: str,
+        *,
+        vector_store_id: str,
+        poll_interval_ms: int | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreFileBatch:
+        """Wait for the given file batch to be processed.
+
+        Note: this will return even if one of the files failed to process, you need to
+        check batch.file_counts.failed_count to handle this case.
+        """
+        headers: dict[str, str] = {"X-Stainless-Poll-Helper": "true"}
+        if is_given(poll_interval_ms):
+            headers["X-Stainless-Custom-Poll-Interval"] = str(poll_interval_ms)
+
+        while True:
+            response = await self.with_raw_response.retrieve(
+                batch_id,
+                vector_store_id=vector_store_id,
+                extra_headers=headers,
+            )
+
+            batch = response.parse()
+            if batch.file_counts.in_progress > 0:
+                if not is_given(poll_interval_ms):
+                    from_header = response.headers.get("openai-poll-after-ms")
+                    if from_header is not None:
+                        poll_interval_ms = int(from_header)
+                    else:
+                        poll_interval_ms = 1000
+
+                await self._sleep(poll_interval_ms / 1000)
+                continue
+
+            return batch
+
+    async def upload_and_poll(
+        self,
+        vector_store_id: str,
+        *,
+        files: Iterable[FileTypes],
+        max_concurrency: int = 5,
+        file_ids: List[str] = [],
+        poll_interval_ms: int | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreFileBatch:
+        """Uploads the given files concurrently and then creates a vector store file batch.
+
+        If you've already uploaded certain files that you want to include in this batch
+        then you can pass their IDs through the `file_ids` argument.
+
+        By default, if any file upload fails then an exception will be eagerly raised.
+
+        The number of concurrency uploads is configurable using the `max_concurrency`
+        parameter.
+
+        Note: this method only supports `asyncio` or `trio` as the backing async
+        runtime.
+        """
+        uploaded_files: list[FileObject] = []
+
+        async_library = sniffio.current_async_library()
+
+        if async_library == "asyncio":
+
+            async def asyncio_upload_file(semaphore: asyncio.Semaphore, file: FileTypes) -> None:
+                async with semaphore:
+                    file_obj = await self._client.files.create(
+                        file=file,
+                        purpose="assistants",
+                    )
+                    uploaded_files.append(file_obj)
+
+            semaphore = asyncio.Semaphore(max_concurrency)
+
+            tasks = [asyncio_upload_file(semaphore, file) for file in files]
+
+            await asyncio.gather(*tasks)
+        elif async_library == "trio":
+            # We only import if the library is being used.
+            # We support Python 3.7 so are using an older version of trio that does not have type information
+            import trio  # type: ignore # pyright: ignore[reportMissingTypeStubs]
+
+            async def trio_upload_file(limiter: trio.CapacityLimiter, file: FileTypes) -> None:
+                async with limiter:
+                    file_obj = await self._client.files.create(
+                        file=file,
+                        purpose="assistants",
+                    )
+                    uploaded_files.append(file_obj)
+
+            limiter = trio.CapacityLimiter(max_concurrency)
+
+            async with trio.open_nursery() as nursery:
+                for file in files:
+                    nursery.start_soon(trio_upload_file, limiter, file)  # pyright: ignore [reportUnknownMemberType]
+        else:
+            raise RuntimeError(
+                f"Async runtime {async_library} is not supported yet. Only asyncio or trio is supported",
+            )
+
+        batch = await self.create_and_poll(
+            vector_store_id=vector_store_id,
+            file_ids=[*file_ids, *(f.id for f in uploaded_files)],
+            poll_interval_ms=poll_interval_ms,
+            chunking_strategy=chunking_strategy,
+        )
+        return batch
+
+
+class FileBatchesWithRawResponse:
+    def __init__(self, file_batches: FileBatches) -> None:
+        self._file_batches = file_batches
+
+        self.create = _legacy_response.to_raw_response_wrapper(
+            file_batches.create,
+        )
+        self.retrieve = _legacy_response.to_raw_response_wrapper(
+            file_batches.retrieve,
+        )
+        self.cancel = _legacy_response.to_raw_response_wrapper(
+            file_batches.cancel,
+        )
+        self.list_files = _legacy_response.to_raw_response_wrapper(
+            file_batches.list_files,
+        )
+
+
+class AsyncFileBatchesWithRawResponse:
+    def __init__(self, file_batches: AsyncFileBatches) -> None:
+        self._file_batches = file_batches
+
+        self.create = _legacy_response.async_to_raw_response_wrapper(
+            file_batches.create,
+        )
+        self.retrieve = _legacy_response.async_to_raw_response_wrapper(
+            file_batches.retrieve,
+        )
+        self.cancel = _legacy_response.async_to_raw_response_wrapper(
+            file_batches.cancel,
+        )
+        self.list_files = _legacy_response.async_to_raw_response_wrapper(
+            file_batches.list_files,
+        )
+
+
+class FileBatchesWithStreamingResponse:
+    def __init__(self, file_batches: FileBatches) -> None:
+        self._file_batches = file_batches
+
+        self.create = to_streamed_response_wrapper(
+            file_batches.create,
+        )
+        self.retrieve = to_streamed_response_wrapper(
+            file_batches.retrieve,
+        )
+        self.cancel = to_streamed_response_wrapper(
+            file_batches.cancel,
+        )
+        self.list_files = to_streamed_response_wrapper(
+            file_batches.list_files,
+        )
+
+
+class AsyncFileBatchesWithStreamingResponse:
+    def __init__(self, file_batches: AsyncFileBatches) -> None:
+        self._file_batches = file_batches
+
+        self.create = async_to_streamed_response_wrapper(
+            file_batches.create,
+        )
+        self.retrieve = async_to_streamed_response_wrapper(
+            file_batches.retrieve,
+        )
+        self.cancel = async_to_streamed_response_wrapper(
+            file_batches.cancel,
+        )
+        self.list_files = async_to_streamed_response_wrapper(
+            file_batches.list_files,
+        )

backend/venv/lib/python3.12/site-packages/openai/resources/beta/vector_stores/files.py

@@ -0,0 +1,726 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+from typing_extensions import Literal, assert_never
+
+import httpx
+
+from .... import _legacy_response
+from ...._types import NOT_GIVEN, Body, Query, Headers, NotGiven, FileTypes
+from ...._utils import (
+    is_given,
+    maybe_transform,
+    async_maybe_transform,
+)
+from ...._compat import cached_property
+from ...._resource import SyncAPIResource, AsyncAPIResource
+from ...._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
+from ....pagination import SyncCursorPage, AsyncCursorPage
+from ....types.beta import FileChunkingStrategyParam
+from ...._base_client import AsyncPaginator, make_request_options
+from ....types.beta.vector_stores import file_list_params, file_create_params
+from ....types.beta.file_chunking_strategy_param import FileChunkingStrategyParam
+from ....types.beta.vector_stores.vector_store_file import VectorStoreFile
+from ....types.beta.vector_stores.vector_store_file_deleted import VectorStoreFileDeleted
+
+__all__ = ["Files", "AsyncFiles"]
+
+
+class Files(SyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> FilesWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return FilesWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> FilesWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return FilesWithStreamingResponse(self)
+
+    def create(
+        self,
+        vector_store_id: str,
+        *,
+        file_id: str,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreFile:
+        """
+        Create a vector store file by attaching a
+        [File](https://platform.openai.com/docs/api-reference/files) to a
+        [vector store](https://platform.openai.com/docs/api-reference/vector-stores/object).
+
+        Args:
+          file_id: A [File](https://platform.openai.com/docs/api-reference/files) ID that the
+              vector store should use. Useful for tools like `file_search` that can access
+              files.
+
+          chunking_strategy: The chunking strategy used to chunk the file(s). If not set, will use the `auto`
+              strategy. Only applicable if `file_ids` is non-empty.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not vector_store_id:
+            raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._post(
+            f"/vector_stores/{vector_store_id}/files",
+            body=maybe_transform(
+                {
+                    "file_id": file_id,
+                    "chunking_strategy": chunking_strategy,
+                },
+                file_create_params.FileCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=VectorStoreFile,
+        )
+
+    def retrieve(
+        self,
+        file_id: str,
+        *,
+        vector_store_id: str,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreFile:
+        """
+        Retrieves a vector store file.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not vector_store_id:
+            raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}")
+        if not file_id:
+            raise ValueError(f"Expected a non-empty value for `file_id` but received {file_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._get(
+            f"/vector_stores/{vector_store_id}/files/{file_id}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=VectorStoreFile,
+        )
+
+    def list(
+        self,
+        vector_store_id: str,
+        *,
+        after: str | NotGiven = NOT_GIVEN,
+        before: str | NotGiven = NOT_GIVEN,
+        filter: Literal["in_progress", "completed", "failed", "cancelled"] | NotGiven = NOT_GIVEN,
+        limit: int | NotGiven = NOT_GIVEN,
+        order: Literal["asc", "desc"] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> SyncCursorPage[VectorStoreFile]:
+        """
+        Returns a list of vector store files.
+
+        Args:
+          after: A cursor for use in pagination. `after` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include after=obj_foo in order to
+              fetch the next page of the list.
+
+          before: A cursor for use in pagination. `before` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include before=obj_foo in order to
+              fetch the previous page of the list.
+
+          filter: Filter by file status. One of `in_progress`, `completed`, `failed`, `cancelled`.
+
+          limit: A limit on the number of objects to be returned. Limit can range between 1 and
+              100, and the default is 20.
+
+          order: Sort order by the `created_at` timestamp of the objects. `asc` for ascending
+              order and `desc` for descending order.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not vector_store_id:
+            raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._get_api_list(
+            f"/vector_stores/{vector_store_id}/files",
+            page=SyncCursorPage[VectorStoreFile],
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform(
+                    {
+                        "after": after,
+                        "before": before,
+                        "filter": filter,
+                        "limit": limit,
+                        "order": order,
+                    },
+                    file_list_params.FileListParams,
+                ),
+            ),
+            model=VectorStoreFile,
+        )
+
+    def delete(
+        self,
+        file_id: str,
+        *,
+        vector_store_id: str,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreFileDeleted:
+        """Delete a vector store file.
+
+        This will remove the file from the vector store but
+        the file itself will not be deleted. To delete the file, use the
+        [delete file](https://platform.openai.com/docs/api-reference/files/delete)
+        endpoint.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not vector_store_id:
+            raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}")
+        if not file_id:
+            raise ValueError(f"Expected a non-empty value for `file_id` but received {file_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._delete(
+            f"/vector_stores/{vector_store_id}/files/{file_id}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=VectorStoreFileDeleted,
+        )
+
+    def create_and_poll(
+        self,
+        file_id: str,
+        *,
+        vector_store_id: str,
+        poll_interval_ms: int | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreFile:
+        """Attach a file to the given vector store and wait for it to be processed."""
+        self.create(vector_store_id=vector_store_id, file_id=file_id, chunking_strategy=chunking_strategy)
+
+        return self.poll(
+            file_id,
+            vector_store_id=vector_store_id,
+            poll_interval_ms=poll_interval_ms,
+        )
+
+    def poll(
+        self,
+        file_id: str,
+        *,
+        vector_store_id: str,
+        poll_interval_ms: int | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreFile:
+        """Wait for the vector store file to finish processing.
+
+        Note: this will return even if the file failed to process, you need to check
+        file.last_error and file.status to handle these cases
+        """
+        headers: dict[str, str] = {"X-Stainless-Poll-Helper": "true"}
+        if is_given(poll_interval_ms):
+            headers["X-Stainless-Custom-Poll-Interval"] = str(poll_interval_ms)
+
+        while True:
+            response = self.with_raw_response.retrieve(
+                file_id,
+                vector_store_id=vector_store_id,
+                extra_headers=headers,
+            )
+
+            file = response.parse()
+            if file.status == "in_progress":
+                if not is_given(poll_interval_ms):
+                    from_header = response.headers.get("openai-poll-after-ms")
+                    if from_header is not None:
+                        poll_interval_ms = int(from_header)
+                    else:
+                        poll_interval_ms = 1000
+
+                self._sleep(poll_interval_ms / 1000)
+            elif file.status == "cancelled" or file.status == "completed" or file.status == "failed":
+                return file
+            else:
+                if TYPE_CHECKING:  # type: ignore[unreachable]
+                    assert_never(file.status)
+                else:
+                    return file
+
+    def upload(
+        self,
+        *,
+        vector_store_id: str,
+        file: FileTypes,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreFile:
+        """Upload a file to the `files` API and then attach it to the given vector store.
+
+        Note the file will be asynchronously processed (you can use the alternative
+        polling helper method to wait for processing to complete).
+        """
+        file_obj = self._client.files.create(file=file, purpose="assistants")
+        return self.create(vector_store_id=vector_store_id, file_id=file_obj.id, chunking_strategy=chunking_strategy)
+
+    def upload_and_poll(
+        self,
+        *,
+        vector_store_id: str,
+        file: FileTypes,
+        poll_interval_ms: int | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreFile:
+        """Add a file to a vector store and poll until processing is complete."""
+        file_obj = self._client.files.create(file=file, purpose="assistants")
+        return self.create_and_poll(
+            vector_store_id=vector_store_id,
+            file_id=file_obj.id,
+            chunking_strategy=chunking_strategy,
+            poll_interval_ms=poll_interval_ms,
+        )
+
+
+class AsyncFiles(AsyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> AsyncFilesWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncFilesWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncFilesWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncFilesWithStreamingResponse(self)
+
+    async def create(
+        self,
+        vector_store_id: str,
+        *,
+        file_id: str,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreFile:
+        """
+        Create a vector store file by attaching a
+        [File](https://platform.openai.com/docs/api-reference/files) to a
+        [vector store](https://platform.openai.com/docs/api-reference/vector-stores/object).
+
+        Args:
+          file_id: A [File](https://platform.openai.com/docs/api-reference/files) ID that the
+              vector store should use. Useful for tools like `file_search` that can access
+              files.
+
+          chunking_strategy: The chunking strategy used to chunk the file(s). If not set, will use the `auto`
+              strategy. Only applicable if `file_ids` is non-empty.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not vector_store_id:
+            raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return await self._post(
+            f"/vector_stores/{vector_store_id}/files",
+            body=await async_maybe_transform(
+                {
+                    "file_id": file_id,
+                    "chunking_strategy": chunking_strategy,
+                },
+                file_create_params.FileCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=VectorStoreFile,
+        )
+
+    async def retrieve(
+        self,
+        file_id: str,
+        *,
+        vector_store_id: str,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreFile:
+        """
+        Retrieves a vector store file.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not vector_store_id:
+            raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}")
+        if not file_id:
+            raise ValueError(f"Expected a non-empty value for `file_id` but received {file_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return await self._get(
+            f"/vector_stores/{vector_store_id}/files/{file_id}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=VectorStoreFile,
+        )
+
+    def list(
+        self,
+        vector_store_id: str,
+        *,
+        after: str | NotGiven = NOT_GIVEN,
+        before: str | NotGiven = NOT_GIVEN,
+        filter: Literal["in_progress", "completed", "failed", "cancelled"] | NotGiven = NOT_GIVEN,
+        limit: int | NotGiven = NOT_GIVEN,
+        order: Literal["asc", "desc"] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> AsyncPaginator[VectorStoreFile, AsyncCursorPage[VectorStoreFile]]:
+        """
+        Returns a list of vector store files.
+
+        Args:
+          after: A cursor for use in pagination. `after` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include after=obj_foo in order to
+              fetch the next page of the list.
+
+          before: A cursor for use in pagination. `before` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include before=obj_foo in order to
+              fetch the previous page of the list.
+
+          filter: Filter by file status. One of `in_progress`, `completed`, `failed`, `cancelled`.
+
+          limit: A limit on the number of objects to be returned. Limit can range between 1 and
+              100, and the default is 20.
+
+          order: Sort order by the `created_at` timestamp of the objects. `asc` for ascending
+              order and `desc` for descending order.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not vector_store_id:
+            raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._get_api_list(
+            f"/vector_stores/{vector_store_id}/files",
+            page=AsyncCursorPage[VectorStoreFile],
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform(
+                    {
+                        "after": after,
+                        "before": before,
+                        "filter": filter,
+                        "limit": limit,
+                        "order": order,
+                    },
+                    file_list_params.FileListParams,
+                ),
+            ),
+            model=VectorStoreFile,
+        )
+
+    async def delete(
+        self,
+        file_id: str,
+        *,
+        vector_store_id: str,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreFileDeleted:
+        """Delete a vector store file.
+
+        This will remove the file from the vector store but
+        the file itself will not be deleted. To delete the file, use the
+        [delete file](https://platform.openai.com/docs/api-reference/files/delete)
+        endpoint.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not vector_store_id:
+            raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}")
+        if not file_id:
+            raise ValueError(f"Expected a non-empty value for `file_id` but received {file_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return await self._delete(
+            f"/vector_stores/{vector_store_id}/files/{file_id}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=VectorStoreFileDeleted,
+        )
+
+    async def create_and_poll(
+        self,
+        file_id: str,
+        *,
+        vector_store_id: str,
+        poll_interval_ms: int | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreFile:
+        """Attach a file to the given vector store and wait for it to be processed."""
+        await self.create(vector_store_id=vector_store_id, file_id=file_id, chunking_strategy=chunking_strategy)
+
+        return await self.poll(
+            file_id,
+            vector_store_id=vector_store_id,
+            poll_interval_ms=poll_interval_ms,
+        )
+
+    async def poll(
+        self,
+        file_id: str,
+        *,
+        vector_store_id: str,
+        poll_interval_ms: int | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreFile:
+        """Wait for the vector store file to finish processing.
+
+        Note: this will return even if the file failed to process, you need to check
+        file.last_error and file.status to handle these cases
+        """
+        headers: dict[str, str] = {"X-Stainless-Poll-Helper": "true"}
+        if is_given(poll_interval_ms):
+            headers["X-Stainless-Custom-Poll-Interval"] = str(poll_interval_ms)
+
+        while True:
+            response = await self.with_raw_response.retrieve(
+                file_id,
+                vector_store_id=vector_store_id,
+                extra_headers=headers,
+            )
+
+            file = response.parse()
+            if file.status == "in_progress":
+                if not is_given(poll_interval_ms):
+                    from_header = response.headers.get("openai-poll-after-ms")
+                    if from_header is not None:
+                        poll_interval_ms = int(from_header)
+                    else:
+                        poll_interval_ms = 1000
+
+                await self._sleep(poll_interval_ms / 1000)
+            elif file.status == "cancelled" or file.status == "completed" or file.status == "failed":
+                return file
+            else:
+                if TYPE_CHECKING:  # type: ignore[unreachable]
+                    assert_never(file.status)
+                else:
+                    return file
+
+    async def upload(
+        self,
+        *,
+        vector_store_id: str,
+        file: FileTypes,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreFile:
+        """Upload a file to the `files` API and then attach it to the given vector store.
+
+        Note the file will be asynchronously processed (you can use the alternative
+        polling helper method to wait for processing to complete).
+        """
+        file_obj = await self._client.files.create(file=file, purpose="assistants")
+        return await self.create(
+            vector_store_id=vector_store_id, file_id=file_obj.id, chunking_strategy=chunking_strategy
+        )
+
+    async def upload_and_poll(
+        self,
+        *,
+        vector_store_id: str,
+        file: FileTypes,
+        poll_interval_ms: int | NotGiven = NOT_GIVEN,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreFile:
+        """Add a file to a vector store and poll until processing is complete."""
+        file_obj = await self._client.files.create(file=file, purpose="assistants")
+        return await self.create_and_poll(
+            vector_store_id=vector_store_id,
+            file_id=file_obj.id,
+            poll_interval_ms=poll_interval_ms,
+            chunking_strategy=chunking_strategy,
+        )
+
+
+class FilesWithRawResponse:
+    def __init__(self, files: Files) -> None:
+        self._files = files
+
+        self.create = _legacy_response.to_raw_response_wrapper(
+            files.create,
+        )
+        self.retrieve = _legacy_response.to_raw_response_wrapper(
+            files.retrieve,
+        )
+        self.list = _legacy_response.to_raw_response_wrapper(
+            files.list,
+        )
+        self.delete = _legacy_response.to_raw_response_wrapper(
+            files.delete,
+        )
+
+
+class AsyncFilesWithRawResponse:
+    def __init__(self, files: AsyncFiles) -> None:
+        self._files = files
+
+        self.create = _legacy_response.async_to_raw_response_wrapper(
+            files.create,
+        )
+        self.retrieve = _legacy_response.async_to_raw_response_wrapper(
+            files.retrieve,
+        )
+        self.list = _legacy_response.async_to_raw_response_wrapper(
+            files.list,
+        )
+        self.delete = _legacy_response.async_to_raw_response_wrapper(
+            files.delete,
+        )
+
+
+class FilesWithStreamingResponse:
+    def __init__(self, files: Files) -> None:
+        self._files = files
+
+        self.create = to_streamed_response_wrapper(
+            files.create,
+        )
+        self.retrieve = to_streamed_response_wrapper(
+            files.retrieve,
+        )
+        self.list = to_streamed_response_wrapper(
+            files.list,
+        )
+        self.delete = to_streamed_response_wrapper(
+            files.delete,
+        )
+
+
+class AsyncFilesWithStreamingResponse:
+    def __init__(self, files: AsyncFiles) -> None:
+        self._files = files
+
+        self.create = async_to_streamed_response_wrapper(
+            files.create,
+        )
+        self.retrieve = async_to_streamed_response_wrapper(
+            files.retrieve,
+        )
+        self.list = async_to_streamed_response_wrapper(
+            files.list,
+        )
+        self.delete = async_to_streamed_response_wrapper(
+            files.delete,
+        )

backend/venv/lib/python3.12/site-packages/openai/resources/beta/vector_stores/vector_stores.py

@@ -0,0 +1,719 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import List, Optional
+from typing_extensions import Literal
+
+import httpx
+
+from .... import _legacy_response
+from .files import (
+    Files,
+    AsyncFiles,
+    FilesWithRawResponse,
+    AsyncFilesWithRawResponse,
+    FilesWithStreamingResponse,
+    AsyncFilesWithStreamingResponse,
+)
+from ...._types import NOT_GIVEN, Body, Query, Headers, NotGiven
+from ...._utils import (
+    maybe_transform,
+    async_maybe_transform,
+)
+from ...._compat import cached_property
+from ...._resource import SyncAPIResource, AsyncAPIResource
+from ...._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
+from .file_batches import (
+    FileBatches,
+    AsyncFileBatches,
+    FileBatchesWithRawResponse,
+    AsyncFileBatchesWithRawResponse,
+    FileBatchesWithStreamingResponse,
+    AsyncFileBatchesWithStreamingResponse,
+)
+from ....pagination import SyncCursorPage, AsyncCursorPage
+from ....types.beta import (
+    FileChunkingStrategyParam,
+    vector_store_list_params,
+    vector_store_create_params,
+    vector_store_update_params,
+)
+from ...._base_client import AsyncPaginator, make_request_options
+from ....types.beta.vector_store import VectorStore
+from ....types.beta.vector_store_deleted import VectorStoreDeleted
+from ....types.beta.file_chunking_strategy_param import FileChunkingStrategyParam
+
+__all__ = ["VectorStores", "AsyncVectorStores"]
+
+
+class VectorStores(SyncAPIResource):
+    @cached_property
+    def files(self) -> Files:
+        return Files(self._client)
+
+    @cached_property
+    def file_batches(self) -> FileBatches:
+        return FileBatches(self._client)
+
+    @cached_property
+    def with_raw_response(self) -> VectorStoresWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return VectorStoresWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> VectorStoresWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return VectorStoresWithStreamingResponse(self)
+
+    def create(
+        self,
+        *,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
+        expires_after: vector_store_create_params.ExpiresAfter | NotGiven = NOT_GIVEN,
+        file_ids: List[str] | NotGiven = NOT_GIVEN,
+        metadata: Optional[object] | NotGiven = NOT_GIVEN,
+        name: str | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> VectorStore:
+        """
+        Create a vector store.
+
+        Args:
+          chunking_strategy: The chunking strategy used to chunk the file(s). If not set, will use the `auto`
+              strategy. Only applicable if `file_ids` is non-empty.
+
+          expires_after: The expiration policy for a vector store.
+
+          file_ids: A list of [File](https://platform.openai.com/docs/api-reference/files) IDs that
+              the vector store should use. Useful for tools like `file_search` that can access
+              files.
+
+          metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
+              for storing additional information about the object in a structured format. Keys
+              can be a maximum of 64 characters long and values can be a maximum of 512
+              characters long.
+
+          name: The name of the vector store.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._post(
+            "/vector_stores",
+            body=maybe_transform(
+                {
+                    "chunking_strategy": chunking_strategy,
+                    "expires_after": expires_after,
+                    "file_ids": file_ids,
+                    "metadata": metadata,
+                    "name": name,
+                },
+                vector_store_create_params.VectorStoreCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=VectorStore,
+        )
+
+    def retrieve(
+        self,
+        vector_store_id: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> VectorStore:
+        """
+        Retrieves a vector store.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not vector_store_id:
+            raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._get(
+            f"/vector_stores/{vector_store_id}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=VectorStore,
+        )
+
+    def update(
+        self,
+        vector_store_id: str,
+        *,
+        expires_after: Optional[vector_store_update_params.ExpiresAfter] | NotGiven = NOT_GIVEN,
+        metadata: Optional[object] | NotGiven = NOT_GIVEN,
+        name: Optional[str] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> VectorStore:
+        """
+        Modifies a vector store.
+
+        Args:
+          expires_after: The expiration policy for a vector store.
+
+          metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
+              for storing additional information about the object in a structured format. Keys
+              can be a maximum of 64 characters long and values can be a maximum of 512
+              characters long.
+
+          name: The name of the vector store.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not vector_store_id:
+            raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._post(
+            f"/vector_stores/{vector_store_id}",
+            body=maybe_transform(
+                {
+                    "expires_after": expires_after,
+                    "metadata": metadata,
+                    "name": name,
+                },
+                vector_store_update_params.VectorStoreUpdateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=VectorStore,
+        )
+
+    def list(
+        self,
+        *,
+        after: str | NotGiven = NOT_GIVEN,
+        before: str | NotGiven = NOT_GIVEN,
+        limit: int | NotGiven = NOT_GIVEN,
+        order: Literal["asc", "desc"] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> SyncCursorPage[VectorStore]:
+        """Returns a list of vector stores.
+
+        Args:
+          after: A cursor for use in pagination.
+
+        `after` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include after=obj_foo in order to
+              fetch the next page of the list.
+
+          before: A cursor for use in pagination. `before` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include before=obj_foo in order to
+              fetch the previous page of the list.
+
+          limit: A limit on the number of objects to be returned. Limit can range between 1 and
+              100, and the default is 20.
+
+          order: Sort order by the `created_at` timestamp of the objects. `asc` for ascending
+              order and `desc` for descending order.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._get_api_list(
+            "/vector_stores",
+            page=SyncCursorPage[VectorStore],
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform(
+                    {
+                        "after": after,
+                        "before": before,
+                        "limit": limit,
+                        "order": order,
+                    },
+                    vector_store_list_params.VectorStoreListParams,
+                ),
+            ),
+            model=VectorStore,
+        )
+
+    def delete(
+        self,
+        vector_store_id: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreDeleted:
+        """
+        Delete a vector store.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not vector_store_id:
+            raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._delete(
+            f"/vector_stores/{vector_store_id}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=VectorStoreDeleted,
+        )
+
+
+class AsyncVectorStores(AsyncAPIResource):
+    @cached_property
+    def files(self) -> AsyncFiles:
+        return AsyncFiles(self._client)
+
+    @cached_property
+    def file_batches(self) -> AsyncFileBatches:
+        return AsyncFileBatches(self._client)
+
+    @cached_property
+    def with_raw_response(self) -> AsyncVectorStoresWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncVectorStoresWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncVectorStoresWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncVectorStoresWithStreamingResponse(self)
+
+    async def create(
+        self,
+        *,
+        chunking_strategy: FileChunkingStrategyParam | NotGiven = NOT_GIVEN,
+        expires_after: vector_store_create_params.ExpiresAfter | NotGiven = NOT_GIVEN,
+        file_ids: List[str] | NotGiven = NOT_GIVEN,
+        metadata: Optional[object] | NotGiven = NOT_GIVEN,
+        name: str | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> VectorStore:
+        """
+        Create a vector store.
+
+        Args:
+          chunking_strategy: The chunking strategy used to chunk the file(s). If not set, will use the `auto`
+              strategy. Only applicable if `file_ids` is non-empty.
+
+          expires_after: The expiration policy for a vector store.
+
+          file_ids: A list of [File](https://platform.openai.com/docs/api-reference/files) IDs that
+              the vector store should use. Useful for tools like `file_search` that can access
+              files.
+
+          metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
+              for storing additional information about the object in a structured format. Keys
+              can be a maximum of 64 characters long and values can be a maximum of 512
+              characters long.
+
+          name: The name of the vector store.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return await self._post(
+            "/vector_stores",
+            body=await async_maybe_transform(
+                {
+                    "chunking_strategy": chunking_strategy,
+                    "expires_after": expires_after,
+                    "file_ids": file_ids,
+                    "metadata": metadata,
+                    "name": name,
+                },
+                vector_store_create_params.VectorStoreCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=VectorStore,
+        )
+
+    async def retrieve(
+        self,
+        vector_store_id: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> VectorStore:
+        """
+        Retrieves a vector store.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not vector_store_id:
+            raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return await self._get(
+            f"/vector_stores/{vector_store_id}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=VectorStore,
+        )
+
+    async def update(
+        self,
+        vector_store_id: str,
+        *,
+        expires_after: Optional[vector_store_update_params.ExpiresAfter] | NotGiven = NOT_GIVEN,
+        metadata: Optional[object] | NotGiven = NOT_GIVEN,
+        name: Optional[str] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> VectorStore:
+        """
+        Modifies a vector store.
+
+        Args:
+          expires_after: The expiration policy for a vector store.
+
+          metadata: Set of 16 key-value pairs that can be attached to an object. This can be useful
+              for storing additional information about the object in a structured format. Keys
+              can be a maximum of 64 characters long and values can be a maximum of 512
+              characters long.
+
+          name: The name of the vector store.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not vector_store_id:
+            raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return await self._post(
+            f"/vector_stores/{vector_store_id}",
+            body=await async_maybe_transform(
+                {
+                    "expires_after": expires_after,
+                    "metadata": metadata,
+                    "name": name,
+                },
+                vector_store_update_params.VectorStoreUpdateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=VectorStore,
+        )
+
+    def list(
+        self,
+        *,
+        after: str | NotGiven = NOT_GIVEN,
+        before: str | NotGiven = NOT_GIVEN,
+        limit: int | NotGiven = NOT_GIVEN,
+        order: Literal["asc", "desc"] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> AsyncPaginator[VectorStore, AsyncCursorPage[VectorStore]]:
+        """Returns a list of vector stores.
+
+        Args:
+          after: A cursor for use in pagination.
+
+        `after` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include after=obj_foo in order to
+              fetch the next page of the list.
+
+          before: A cursor for use in pagination. `before` is an object ID that defines your place
+              in the list. For instance, if you make a list request and receive 100 objects,
+              ending with obj_foo, your subsequent call can include before=obj_foo in order to
+              fetch the previous page of the list.
+
+          limit: A limit on the number of objects to be returned. Limit can range between 1 and
+              100, and the default is 20.
+
+          order: Sort order by the `created_at` timestamp of the objects. `asc` for ascending
+              order and `desc` for descending order.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._get_api_list(
+            "/vector_stores",
+            page=AsyncCursorPage[VectorStore],
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform(
+                    {
+                        "after": after,
+                        "before": before,
+                        "limit": limit,
+                        "order": order,
+                    },
+                    vector_store_list_params.VectorStoreListParams,
+                ),
+            ),
+            model=VectorStore,
+        )
+
+    async def delete(
+        self,
+        vector_store_id: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> VectorStoreDeleted:
+        """
+        Delete a vector store.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not vector_store_id:
+            raise ValueError(f"Expected a non-empty value for `vector_store_id` but received {vector_store_id!r}")
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return await self._delete(
+            f"/vector_stores/{vector_store_id}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=VectorStoreDeleted,
+        )
+
+
+class VectorStoresWithRawResponse:
+    def __init__(self, vector_stores: VectorStores) -> None:
+        self._vector_stores = vector_stores
+
+        self.create = _legacy_response.to_raw_response_wrapper(
+            vector_stores.create,
+        )
+        self.retrieve = _legacy_response.to_raw_response_wrapper(
+            vector_stores.retrieve,
+        )
+        self.update = _legacy_response.to_raw_response_wrapper(
+            vector_stores.update,
+        )
+        self.list = _legacy_response.to_raw_response_wrapper(
+            vector_stores.list,
+        )
+        self.delete = _legacy_response.to_raw_response_wrapper(
+            vector_stores.delete,
+        )
+
+    @cached_property
+    def files(self) -> FilesWithRawResponse:
+        return FilesWithRawResponse(self._vector_stores.files)
+
+    @cached_property
+    def file_batches(self) -> FileBatchesWithRawResponse:
+        return FileBatchesWithRawResponse(self._vector_stores.file_batches)
+
+
+class AsyncVectorStoresWithRawResponse:
+    def __init__(self, vector_stores: AsyncVectorStores) -> None:
+        self._vector_stores = vector_stores
+
+        self.create = _legacy_response.async_to_raw_response_wrapper(
+            vector_stores.create,
+        )
+        self.retrieve = _legacy_response.async_to_raw_response_wrapper(
+            vector_stores.retrieve,
+        )
+        self.update = _legacy_response.async_to_raw_response_wrapper(
+            vector_stores.update,
+        )
+        self.list = _legacy_response.async_to_raw_response_wrapper(
+            vector_stores.list,
+        )
+        self.delete = _legacy_response.async_to_raw_response_wrapper(
+            vector_stores.delete,
+        )
+
+    @cached_property
+    def files(self) -> AsyncFilesWithRawResponse:
+        return AsyncFilesWithRawResponse(self._vector_stores.files)
+
+    @cached_property
+    def file_batches(self) -> AsyncFileBatchesWithRawResponse:
+        return AsyncFileBatchesWithRawResponse(self._vector_stores.file_batches)
+
+
+class VectorStoresWithStreamingResponse:
+    def __init__(self, vector_stores: VectorStores) -> None:
+        self._vector_stores = vector_stores
+
+        self.create = to_streamed_response_wrapper(
+            vector_stores.create,
+        )
+        self.retrieve = to_streamed_response_wrapper(
+            vector_stores.retrieve,
+        )
+        self.update = to_streamed_response_wrapper(
+            vector_stores.update,
+        )
+        self.list = to_streamed_response_wrapper(
+            vector_stores.list,
+        )
+        self.delete = to_streamed_response_wrapper(
+            vector_stores.delete,
+        )
+
+    @cached_property
+    def files(self) -> FilesWithStreamingResponse:
+        return FilesWithStreamingResponse(self._vector_stores.files)
+
+    @cached_property
+    def file_batches(self) -> FileBatchesWithStreamingResponse:
+        return FileBatchesWithStreamingResponse(self._vector_stores.file_batches)
+
+
+class AsyncVectorStoresWithStreamingResponse:
+    def __init__(self, vector_stores: AsyncVectorStores) -> None:
+        self._vector_stores = vector_stores
+
+        self.create = async_to_streamed_response_wrapper(
+            vector_stores.create,
+        )
+        self.retrieve = async_to_streamed_response_wrapper(
+            vector_stores.retrieve,
+        )
+        self.update = async_to_streamed_response_wrapper(
+            vector_stores.update,
+        )
+        self.list = async_to_streamed_response_wrapper(
+            vector_stores.list,
+        )
+        self.delete = async_to_streamed_response_wrapper(
+            vector_stores.delete,
+        )
+
+    @cached_property
+    def files(self) -> AsyncFilesWithStreamingResponse:
+        return AsyncFilesWithStreamingResponse(self._vector_stores.files)
+
+    @cached_property
+    def file_batches(self) -> AsyncFileBatchesWithStreamingResponse:
+        return AsyncFileBatchesWithStreamingResponse(self._vector_stores.file_batches)

backend/venv/lib/python3.12/site-packages/openai/resources/chat/__init__.py

@@ -0,0 +1,33 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from .chat import (
+    Chat,
+    AsyncChat,
+    ChatWithRawResponse,
+    AsyncChatWithRawResponse,
+    ChatWithStreamingResponse,
+    AsyncChatWithStreamingResponse,
+)
+from .completions import (
+    Completions,
+    AsyncCompletions,
+    CompletionsWithRawResponse,
+    AsyncCompletionsWithRawResponse,
+    CompletionsWithStreamingResponse,
+    AsyncCompletionsWithStreamingResponse,
+)
+
+__all__ = [
+    "Completions",
+    "AsyncCompletions",
+    "CompletionsWithRawResponse",
+    "AsyncCompletionsWithRawResponse",
+    "CompletionsWithStreamingResponse",
+    "AsyncCompletionsWithStreamingResponse",
+    "Chat",
+    "AsyncChat",
+    "ChatWithRawResponse",
+    "AsyncChatWithRawResponse",
+    "ChatWithStreamingResponse",
+    "AsyncChatWithStreamingResponse",
+]

backend/venv/lib/python3.12/site-packages/openai/resources/chat/chat.py

@@ -0,0 +1,102 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from ..._compat import cached_property
+from ..._resource import SyncAPIResource, AsyncAPIResource
+from .completions import (
+    Completions,
+    AsyncCompletions,
+    CompletionsWithRawResponse,
+    AsyncCompletionsWithRawResponse,
+    CompletionsWithStreamingResponse,
+    AsyncCompletionsWithStreamingResponse,
+)
+
+__all__ = ["Chat", "AsyncChat"]
+
+
+class Chat(SyncAPIResource):
+    @cached_property
+    def completions(self) -> Completions:
+        return Completions(self._client)
+
+    @cached_property
+    def with_raw_response(self) -> ChatWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return ChatWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> ChatWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return ChatWithStreamingResponse(self)
+
+
+class AsyncChat(AsyncAPIResource):
+    @cached_property
+    def completions(self) -> AsyncCompletions:
+        return AsyncCompletions(self._client)
+
+    @cached_property
+    def with_raw_response(self) -> AsyncChatWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncChatWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncChatWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncChatWithStreamingResponse(self)
+
+
+class ChatWithRawResponse:
+    def __init__(self, chat: Chat) -> None:
+        self._chat = chat
+
+    @cached_property
+    def completions(self) -> CompletionsWithRawResponse:
+        return CompletionsWithRawResponse(self._chat.completions)
+
+
+class AsyncChatWithRawResponse:
+    def __init__(self, chat: AsyncChat) -> None:
+        self._chat = chat
+
+    @cached_property
+    def completions(self) -> AsyncCompletionsWithRawResponse:
+        return AsyncCompletionsWithRawResponse(self._chat.completions)
+
+
+class ChatWithStreamingResponse:
+    def __init__(self, chat: Chat) -> None:
+        self._chat = chat
+
+    @cached_property
+    def completions(self) -> CompletionsWithStreamingResponse:
+        return CompletionsWithStreamingResponse(self._chat.completions)
+
+
+class AsyncChatWithStreamingResponse:
+    def __init__(self, chat: AsyncChat) -> None:
+        self._chat = chat
+
+    @cached_property
+    def completions(self) -> AsyncCompletionsWithStreamingResponse:
+        return AsyncCompletionsWithStreamingResponse(self._chat.completions)

backend/venv/lib/python3.12/site-packages/openai/resources/embeddings.py

@@ -0,0 +1,283 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+import base64
+from typing import List, Union, Iterable, cast
+from typing_extensions import Literal
+
+import httpx
+
+from .. import _legacy_response
+from ..types import embedding_create_params
+from .._types import NOT_GIVEN, Body, Query, Headers, NotGiven
+from .._utils import is_given, maybe_transform
+from .._compat import cached_property
+from .._extras import numpy as np, has_numpy
+from .._resource import SyncAPIResource, AsyncAPIResource
+from .._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
+from .._base_client import make_request_options
+from ..types.embedding_model import EmbeddingModel
+from ..types.create_embedding_response import CreateEmbeddingResponse
+
+__all__ = ["Embeddings", "AsyncEmbeddings"]
+
+
+class Embeddings(SyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> EmbeddingsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return EmbeddingsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> EmbeddingsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return EmbeddingsWithStreamingResponse(self)
+
+    def create(
+        self,
+        *,
+        input: Union[str, List[str], Iterable[int], Iterable[Iterable[int]]],
+        model: Union[str, EmbeddingModel],
+        dimensions: int | NotGiven = NOT_GIVEN,
+        encoding_format: Literal["float", "base64"] | NotGiven = NOT_GIVEN,
+        user: str | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> CreateEmbeddingResponse:
+        """
+        Creates an embedding vector representing the input text.
+
+        Args:
+          input: Input text to embed, encoded as a string or array of tokens. To embed multiple
+              inputs in a single request, pass an array of strings or array of token arrays.
+              The input must not exceed the max input tokens for the model (8192 tokens for
+              `text-embedding-ada-002`), cannot be an empty string, and any array must be 2048
+              dimensions or less.
+              [Example Python code](https://cookbook.openai.com/examples/how_to_count_tokens_with_tiktoken)
+              for counting tokens.
+
+          model: ID of the model to use. You can use the
+              [List models](https://platform.openai.com/docs/api-reference/models/list) API to
+              see all of your available models, or see our
+              [Model overview](https://platform.openai.com/docs/models/overview) for
+              descriptions of them.
+
+          dimensions: The number of dimensions the resulting output embeddings should have. Only
+              supported in `text-embedding-3` and later models.
+
+          encoding_format: The format to return the embeddings in. Can be either `float` or
+              [`base64`](https://pypi.org/project/pybase64/).
+
+          user: A unique identifier representing your end-user, which can help OpenAI to monitor
+              and detect abuse.
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        params = {
+            "input": input,
+            "model": model,
+            "user": user,
+            "dimensions": dimensions,
+            "encoding_format": encoding_format,
+        }
+        if not is_given(encoding_format) and has_numpy():
+            params["encoding_format"] = "base64"
+
+        def parser(obj: CreateEmbeddingResponse) -> CreateEmbeddingResponse:
+            if is_given(encoding_format):
+                # don't modify the response object if a user explicitly asked for a format
+                return obj
+
+            for embedding in obj.data:
+                data = cast(object, embedding.embedding)
+                if not isinstance(data, str):
+                    # numpy is not installed / base64 optimisation isn't enabled for this model yet
+                    continue
+
+                embedding.embedding = np.frombuffer(  # type: ignore[no-untyped-call]
+                    base64.b64decode(data), dtype="float32"
+                ).tolist()
+
+            return obj
+
+        return self._post(
+            "/embeddings",
+            body=maybe_transform(params, embedding_create_params.EmbeddingCreateParams),
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                post_parser=parser,
+            ),
+            cast_to=CreateEmbeddingResponse,
+        )
+
+
+class AsyncEmbeddings(AsyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> AsyncEmbeddingsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncEmbeddingsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncEmbeddingsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncEmbeddingsWithStreamingResponse(self)
+
+    async def create(
+        self,
+        *,
+        input: Union[str, List[str], Iterable[int], Iterable[Iterable[int]]],
+        model: Union[str, EmbeddingModel],
+        dimensions: int | NotGiven = NOT_GIVEN,
+        encoding_format: Literal["float", "base64"] | NotGiven = NOT_GIVEN,
+        user: str | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> CreateEmbeddingResponse:
+        """
+        Creates an embedding vector representing the input text.
+
+        Args:
+          input: Input text to embed, encoded as a string or array of tokens. To embed multiple
+              inputs in a single request, pass an array of strings or array of token arrays.
+              The input must not exceed the max input tokens for the model (8192 tokens for
+              `text-embedding-ada-002`), cannot be an empty string, and any array must be 2048
+              dimensions or less.
+              [Example Python code](https://cookbook.openai.com/examples/how_to_count_tokens_with_tiktoken)
+              for counting tokens.
+
+          model: ID of the model to use. You can use the
+              [List models](https://platform.openai.com/docs/api-reference/models/list) API to
+              see all of your available models, or see our
+              [Model overview](https://platform.openai.com/docs/models/overview) for
+              descriptions of them.
+
+          dimensions: The number of dimensions the resulting output embeddings should have. Only
+              supported in `text-embedding-3` and later models.
+
+          encoding_format: The format to return the embeddings in. Can be either `float` or
+              [`base64`](https://pypi.org/project/pybase64/).
+
+          user: A unique identifier representing your end-user, which can help OpenAI to monitor
+              and detect abuse.
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        params = {
+            "input": input,
+            "model": model,
+            "user": user,
+            "dimensions": dimensions,
+            "encoding_format": encoding_format,
+        }
+        if not is_given(encoding_format) and has_numpy():
+            params["encoding_format"] = "base64"
+
+        def parser(obj: CreateEmbeddingResponse) -> CreateEmbeddingResponse:
+            if is_given(encoding_format):
+                # don't modify the response object if a user explicitly asked for a format
+                return obj
+
+            for embedding in obj.data:
+                data = cast(object, embedding.embedding)
+                if not isinstance(data, str):
+                    # numpy is not installed / base64 optimisation isn't enabled for this model yet
+                    continue
+
+                embedding.embedding = np.frombuffer(  # type: ignore[no-untyped-call]
+                    base64.b64decode(data), dtype="float32"
+                ).tolist()
+
+            return obj
+
+        return await self._post(
+            "/embeddings",
+            body=maybe_transform(params, embedding_create_params.EmbeddingCreateParams),
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                post_parser=parser,
+            ),
+            cast_to=CreateEmbeddingResponse,
+        )
+
+
+class EmbeddingsWithRawResponse:
+    def __init__(self, embeddings: Embeddings) -> None:
+        self._embeddings = embeddings
+
+        self.create = _legacy_response.to_raw_response_wrapper(
+            embeddings.create,
+        )
+
+
+class AsyncEmbeddingsWithRawResponse:
+    def __init__(self, embeddings: AsyncEmbeddings) -> None:
+        self._embeddings = embeddings
+
+        self.create = _legacy_response.async_to_raw_response_wrapper(
+            embeddings.create,
+        )
+
+
+class EmbeddingsWithStreamingResponse:
+    def __init__(self, embeddings: Embeddings) -> None:
+        self._embeddings = embeddings
+
+        self.create = to_streamed_response_wrapper(
+            embeddings.create,
+        )
+
+
+class AsyncEmbeddingsWithStreamingResponse:
+    def __init__(self, embeddings: AsyncEmbeddings) -> None:
+        self._embeddings = embeddings
+
+        self.create = async_to_streamed_response_wrapper(
+            embeddings.create,
+        )

backend/venv/lib/python3.12/site-packages/openai/resources/files.py

@@ -0,0 +1,731 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+import time
+import typing_extensions
+from typing import Mapping, cast
+
+import httpx
+
+from .. import _legacy_response
+from ..types import FilePurpose, file_list_params, file_create_params
+from .._types import NOT_GIVEN, Body, Query, Headers, NotGiven, FileTypes
+from .._utils import (
+    extract_files,
+    maybe_transform,
+    deepcopy_minimal,
+    async_maybe_transform,
+)
+from .._compat import cached_property
+from .._resource import SyncAPIResource, AsyncAPIResource
+from .._response import (
+    StreamedBinaryAPIResponse,
+    AsyncStreamedBinaryAPIResponse,
+    to_streamed_response_wrapper,
+    async_to_streamed_response_wrapper,
+    to_custom_streamed_response_wrapper,
+    async_to_custom_streamed_response_wrapper,
+)
+from ..pagination import SyncPage, AsyncPage
+from .._base_client import (
+    AsyncPaginator,
+    make_request_options,
+)
+from ..types.file_object import FileObject
+from ..types.file_deleted import FileDeleted
+from ..types.file_purpose import FilePurpose
+
+__all__ = ["Files", "AsyncFiles"]
+
+
+class Files(SyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> FilesWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return FilesWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> FilesWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return FilesWithStreamingResponse(self)
+
+    def create(
+        self,
+        *,
+        file: FileTypes,
+        purpose: FilePurpose,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> FileObject:
+        """Upload a file that can be used across various endpoints.
+
+        Individual files can be
+        up to 512 MB, and the size of all files uploaded by one organization can be up
+        to 100 GB.
+
+        The Assistants API supports files up to 2 million tokens and of specific file
+        types. See the
+        [Assistants Tools guide](https://platform.openai.com/docs/assistants/tools) for
+        details.
+
+        The Fine-tuning API only supports `.jsonl` files. The input also has certain
+        required formats for fine-tuning
+        [chat](https://platform.openai.com/docs/api-reference/fine-tuning/chat-input) or
+        [completions](https://platform.openai.com/docs/api-reference/fine-tuning/completions-input)
+        models.
+
+        The Batch API only supports `.jsonl` files up to 100 MB in size. The input also
+        has a specific required
+        [format](https://platform.openai.com/docs/api-reference/batch/request-input).
+
+        Please [contact us](https://help.openai.com/) if you need to increase these
+        storage limits.
+
+        Args:
+          file: The File object (not file name) to be uploaded.
+
+          purpose: The intended purpose of the uploaded file.
+
+              Use "assistants" for
+              [Assistants](https://platform.openai.com/docs/api-reference/assistants) and
+              [Message](https://platform.openai.com/docs/api-reference/messages) files,
+              "vision" for Assistants image file inputs, "batch" for
+              [Batch API](https://platform.openai.com/docs/guides/batch), and "fine-tune" for
+              [Fine-tuning](https://platform.openai.com/docs/api-reference/fine-tuning).
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        body = deepcopy_minimal(
+            {
+                "file": file,
+                "purpose": purpose,
+            }
+        )
+        files = extract_files(cast(Mapping[str, object], body), paths=[["file"]])
+        # It should be noted that the actual Content-Type header that will be
+        # sent to the server will contain a `boundary` parameter, e.g.
+        # multipart/form-data; boundary=---abc--
+        extra_headers = {"Content-Type": "multipart/form-data", **(extra_headers or {})}
+        return self._post(
+            "/files",
+            body=maybe_transform(body, file_create_params.FileCreateParams),
+            files=files,
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=FileObject,
+        )
+
+    def retrieve(
+        self,
+        file_id: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> FileObject:
+        """
+        Returns information about a specific file.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not file_id:
+            raise ValueError(f"Expected a non-empty value for `file_id` but received {file_id!r}")
+        return self._get(
+            f"/files/{file_id}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=FileObject,
+        )
+
+    def list(
+        self,
+        *,
+        purpose: str | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> SyncPage[FileObject]:
+        """
+        Returns a list of files that belong to the user's organization.
+
+        Args:
+          purpose: Only return files with the given purpose.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        return self._get_api_list(
+            "/files",
+            page=SyncPage[FileObject],
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform({"purpose": purpose}, file_list_params.FileListParams),
+            ),
+            model=FileObject,
+        )
+
+    def delete(
+        self,
+        file_id: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> FileDeleted:
+        """
+        Delete a file.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not file_id:
+            raise ValueError(f"Expected a non-empty value for `file_id` but received {file_id!r}")
+        return self._delete(
+            f"/files/{file_id}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=FileDeleted,
+        )
+
+    def content(
+        self,
+        file_id: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> _legacy_response.HttpxBinaryResponseContent:
+        """
+        Returns the contents of the specified file.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not file_id:
+            raise ValueError(f"Expected a non-empty value for `file_id` but received {file_id!r}")
+        extra_headers = {"Accept": "application/binary", **(extra_headers or {})}
+        return self._get(
+            f"/files/{file_id}/content",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=_legacy_response.HttpxBinaryResponseContent,
+        )
+
+    @typing_extensions.deprecated("The `.content()` method should be used instead")
+    def retrieve_content(
+        self,
+        file_id: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> str:
+        """
+        Returns the contents of the specified file.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not file_id:
+            raise ValueError(f"Expected a non-empty value for `file_id` but received {file_id!r}")
+        return self._get(
+            f"/files/{file_id}/content",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=str,
+        )
+
+    def wait_for_processing(
+        self,
+        id: str,
+        *,
+        poll_interval: float = 5.0,
+        max_wait_seconds: float = 30 * 60,
+    ) -> FileObject:
+        """Waits for the given file to be processed, default timeout is 30 mins."""
+        TERMINAL_STATES = {"processed", "error", "deleted"}
+
+        start = time.time()
+        file = self.retrieve(id)
+        while file.status not in TERMINAL_STATES:
+            self._sleep(poll_interval)
+
+            file = self.retrieve(id)
+            if time.time() - start > max_wait_seconds:
+                raise RuntimeError(
+                    f"Giving up on waiting for file {id} to finish processing after {max_wait_seconds} seconds."
+                )
+
+        return file
+
+
+class AsyncFiles(AsyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> AsyncFilesWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncFilesWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncFilesWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncFilesWithStreamingResponse(self)
+
+    async def create(
+        self,
+        *,
+        file: FileTypes,
+        purpose: FilePurpose,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> FileObject:
+        """Upload a file that can be used across various endpoints.
+
+        Individual files can be
+        up to 512 MB, and the size of all files uploaded by one organization can be up
+        to 100 GB.
+
+        The Assistants API supports files up to 2 million tokens and of specific file
+        types. See the
+        [Assistants Tools guide](https://platform.openai.com/docs/assistants/tools) for
+        details.
+
+        The Fine-tuning API only supports `.jsonl` files. The input also has certain
+        required formats for fine-tuning
+        [chat](https://platform.openai.com/docs/api-reference/fine-tuning/chat-input) or
+        [completions](https://platform.openai.com/docs/api-reference/fine-tuning/completions-input)
+        models.
+
+        The Batch API only supports `.jsonl` files up to 100 MB in size. The input also
+        has a specific required
+        [format](https://platform.openai.com/docs/api-reference/batch/request-input).
+
+        Please [contact us](https://help.openai.com/) if you need to increase these
+        storage limits.
+
+        Args:
+          file: The File object (not file name) to be uploaded.
+
+          purpose: The intended purpose of the uploaded file.
+
+              Use "assistants" for
+              [Assistants](https://platform.openai.com/docs/api-reference/assistants) and
+              [Message](https://platform.openai.com/docs/api-reference/messages) files,
+              "vision" for Assistants image file inputs, "batch" for
+              [Batch API](https://platform.openai.com/docs/guides/batch), and "fine-tune" for
+              [Fine-tuning](https://platform.openai.com/docs/api-reference/fine-tuning).
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        body = deepcopy_minimal(
+            {
+                "file": file,
+                "purpose": purpose,
+            }
+        )
+        files = extract_files(cast(Mapping[str, object], body), paths=[["file"]])
+        # It should be noted that the actual Content-Type header that will be
+        # sent to the server will contain a `boundary` parameter, e.g.
+        # multipart/form-data; boundary=---abc--
+        extra_headers = {"Content-Type": "multipart/form-data", **(extra_headers or {})}
+        return await self._post(
+            "/files",
+            body=await async_maybe_transform(body, file_create_params.FileCreateParams),
+            files=files,
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=FileObject,
+        )
+
+    async def retrieve(
+        self,
+        file_id: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> FileObject:
+        """
+        Returns information about a specific file.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not file_id:
+            raise ValueError(f"Expected a non-empty value for `file_id` but received {file_id!r}")
+        return await self._get(
+            f"/files/{file_id}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=FileObject,
+        )
+
+    def list(
+        self,
+        *,
+        purpose: str | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> AsyncPaginator[FileObject, AsyncPage[FileObject]]:
+        """
+        Returns a list of files that belong to the user's organization.
+
+        Args:
+          purpose: Only return files with the given purpose.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        return self._get_api_list(
+            "/files",
+            page=AsyncPage[FileObject],
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform({"purpose": purpose}, file_list_params.FileListParams),
+            ),
+            model=FileObject,
+        )
+
+    async def delete(
+        self,
+        file_id: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> FileDeleted:
+        """
+        Delete a file.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not file_id:
+            raise ValueError(f"Expected a non-empty value for `file_id` but received {file_id!r}")
+        return await self._delete(
+            f"/files/{file_id}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=FileDeleted,
+        )
+
+    async def content(
+        self,
+        file_id: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> _legacy_response.HttpxBinaryResponseContent:
+        """
+        Returns the contents of the specified file.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not file_id:
+            raise ValueError(f"Expected a non-empty value for `file_id` but received {file_id!r}")
+        extra_headers = {"Accept": "application/binary", **(extra_headers or {})}
+        return await self._get(
+            f"/files/{file_id}/content",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=_legacy_response.HttpxBinaryResponseContent,
+        )
+
+    @typing_extensions.deprecated("The `.content()` method should be used instead")
+    async def retrieve_content(
+        self,
+        file_id: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> str:
+        """
+        Returns the contents of the specified file.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not file_id:
+            raise ValueError(f"Expected a non-empty value for `file_id` but received {file_id!r}")
+        return await self._get(
+            f"/files/{file_id}/content",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=str,
+        )
+
+    async def wait_for_processing(
+        self,
+        id: str,
+        *,
+        poll_interval: float = 5.0,
+        max_wait_seconds: float = 30 * 60,
+    ) -> FileObject:
+        """Waits for the given file to be processed, default timeout is 30 mins."""
+        TERMINAL_STATES = {"processed", "error", "deleted"}
+
+        start = time.time()
+        file = await self.retrieve(id)
+        while file.status not in TERMINAL_STATES:
+            await self._sleep(poll_interval)
+
+            file = await self.retrieve(id)
+            if time.time() - start > max_wait_seconds:
+                raise RuntimeError(
+                    f"Giving up on waiting for file {id} to finish processing after {max_wait_seconds} seconds."
+                )
+
+        return file
+
+
+class FilesWithRawResponse:
+    def __init__(self, files: Files) -> None:
+        self._files = files
+
+        self.create = _legacy_response.to_raw_response_wrapper(
+            files.create,
+        )
+        self.retrieve = _legacy_response.to_raw_response_wrapper(
+            files.retrieve,
+        )
+        self.list = _legacy_response.to_raw_response_wrapper(
+            files.list,
+        )
+        self.delete = _legacy_response.to_raw_response_wrapper(
+            files.delete,
+        )
+        self.content = _legacy_response.to_raw_response_wrapper(
+            files.content,
+        )
+        self.retrieve_content = (  # pyright: ignore[reportDeprecated]
+            _legacy_response.to_raw_response_wrapper(
+                files.retrieve_content  # pyright: ignore[reportDeprecated],
+            )
+        )
+
+
+class AsyncFilesWithRawResponse:
+    def __init__(self, files: AsyncFiles) -> None:
+        self._files = files
+
+        self.create = _legacy_response.async_to_raw_response_wrapper(
+            files.create,
+        )
+        self.retrieve = _legacy_response.async_to_raw_response_wrapper(
+            files.retrieve,
+        )
+        self.list = _legacy_response.async_to_raw_response_wrapper(
+            files.list,
+        )
+        self.delete = _legacy_response.async_to_raw_response_wrapper(
+            files.delete,
+        )
+        self.content = _legacy_response.async_to_raw_response_wrapper(
+            files.content,
+        )
+        self.retrieve_content = (  # pyright: ignore[reportDeprecated]
+            _legacy_response.async_to_raw_response_wrapper(
+                files.retrieve_content  # pyright: ignore[reportDeprecated],
+            )
+        )
+
+
+class FilesWithStreamingResponse:
+    def __init__(self, files: Files) -> None:
+        self._files = files
+
+        self.create = to_streamed_response_wrapper(
+            files.create,
+        )
+        self.retrieve = to_streamed_response_wrapper(
+            files.retrieve,
+        )
+        self.list = to_streamed_response_wrapper(
+            files.list,
+        )
+        self.delete = to_streamed_response_wrapper(
+            files.delete,
+        )
+        self.content = to_custom_streamed_response_wrapper(
+            files.content,
+            StreamedBinaryAPIResponse,
+        )
+        self.retrieve_content = (  # pyright: ignore[reportDeprecated]
+            to_streamed_response_wrapper(
+                files.retrieve_content  # pyright: ignore[reportDeprecated],
+            )
+        )
+
+
+class AsyncFilesWithStreamingResponse:
+    def __init__(self, files: AsyncFiles) -> None:
+        self._files = files
+
+        self.create = async_to_streamed_response_wrapper(
+            files.create,
+        )
+        self.retrieve = async_to_streamed_response_wrapper(
+            files.retrieve,
+        )
+        self.list = async_to_streamed_response_wrapper(
+            files.list,
+        )
+        self.delete = async_to_streamed_response_wrapper(
+            files.delete,
+        )
+        self.content = async_to_custom_streamed_response_wrapper(
+            files.content,
+            AsyncStreamedBinaryAPIResponse,
+        )
+        self.retrieve_content = (  # pyright: ignore[reportDeprecated]
+            async_to_streamed_response_wrapper(
+                files.retrieve_content  # pyright: ignore[reportDeprecated],
+            )
+        )

backend/venv/lib/python3.12/site-packages/openai/resources/fine_tuning/__init__.py

@@ -0,0 +1,33 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from .jobs import (
+    Jobs,
+    AsyncJobs,
+    JobsWithRawResponse,
+    AsyncJobsWithRawResponse,
+    JobsWithStreamingResponse,
+    AsyncJobsWithStreamingResponse,
+)
+from .fine_tuning import (
+    FineTuning,
+    AsyncFineTuning,
+    FineTuningWithRawResponse,
+    AsyncFineTuningWithRawResponse,
+    FineTuningWithStreamingResponse,
+    AsyncFineTuningWithStreamingResponse,
+)
+
+__all__ = [
+    "Jobs",
+    "AsyncJobs",
+    "JobsWithRawResponse",
+    "AsyncJobsWithRawResponse",
+    "JobsWithStreamingResponse",
+    "AsyncJobsWithStreamingResponse",
+    "FineTuning",
+    "AsyncFineTuning",
+    "FineTuningWithRawResponse",
+    "AsyncFineTuningWithRawResponse",
+    "FineTuningWithStreamingResponse",
+    "AsyncFineTuningWithStreamingResponse",
+]

backend/venv/lib/python3.12/site-packages/openai/resources/fine_tuning/fine_tuning.py

@@ -0,0 +1,103 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from .jobs import (
+    Jobs,
+    AsyncJobs,
+    JobsWithRawResponse,
+    AsyncJobsWithRawResponse,
+    JobsWithStreamingResponse,
+    AsyncJobsWithStreamingResponse,
+)
+from ..._compat import cached_property
+from .jobs.jobs import Jobs, AsyncJobs
+from ..._resource import SyncAPIResource, AsyncAPIResource
+
+__all__ = ["FineTuning", "AsyncFineTuning"]
+
+
+class FineTuning(SyncAPIResource):
+    @cached_property
+    def jobs(self) -> Jobs:
+        return Jobs(self._client)
+
+    @cached_property
+    def with_raw_response(self) -> FineTuningWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return FineTuningWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> FineTuningWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return FineTuningWithStreamingResponse(self)
+
+
+class AsyncFineTuning(AsyncAPIResource):
+    @cached_property
+    def jobs(self) -> AsyncJobs:
+        return AsyncJobs(self._client)
+
+    @cached_property
+    def with_raw_response(self) -> AsyncFineTuningWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncFineTuningWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncFineTuningWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncFineTuningWithStreamingResponse(self)
+
+
+class FineTuningWithRawResponse:
+    def __init__(self, fine_tuning: FineTuning) -> None:
+        self._fine_tuning = fine_tuning
+
+    @cached_property
+    def jobs(self) -> JobsWithRawResponse:
+        return JobsWithRawResponse(self._fine_tuning.jobs)
+
+
+class AsyncFineTuningWithRawResponse:
+    def __init__(self, fine_tuning: AsyncFineTuning) -> None:
+        self._fine_tuning = fine_tuning
+
+    @cached_property
+    def jobs(self) -> AsyncJobsWithRawResponse:
+        return AsyncJobsWithRawResponse(self._fine_tuning.jobs)
+
+
+class FineTuningWithStreamingResponse:
+    def __init__(self, fine_tuning: FineTuning) -> None:
+        self._fine_tuning = fine_tuning
+
+    @cached_property
+    def jobs(self) -> JobsWithStreamingResponse:
+        return JobsWithStreamingResponse(self._fine_tuning.jobs)
+
+
+class AsyncFineTuningWithStreamingResponse:
+    def __init__(self, fine_tuning: AsyncFineTuning) -> None:
+        self._fine_tuning = fine_tuning
+
+    @cached_property
+    def jobs(self) -> AsyncJobsWithStreamingResponse:
+        return AsyncJobsWithStreamingResponse(self._fine_tuning.jobs)

backend/venv/lib/python3.12/site-packages/openai/resources/fine_tuning/jobs/__init__.py

@@ -0,0 +1,33 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from .jobs import (
+    Jobs,
+    AsyncJobs,
+    JobsWithRawResponse,
+    AsyncJobsWithRawResponse,
+    JobsWithStreamingResponse,
+    AsyncJobsWithStreamingResponse,
+)
+from .checkpoints import (
+    Checkpoints,
+    AsyncCheckpoints,
+    CheckpointsWithRawResponse,
+    AsyncCheckpointsWithRawResponse,
+    CheckpointsWithStreamingResponse,
+    AsyncCheckpointsWithStreamingResponse,
+)
+
+__all__ = [
+    "Checkpoints",
+    "AsyncCheckpoints",
+    "CheckpointsWithRawResponse",
+    "AsyncCheckpointsWithRawResponse",
+    "CheckpointsWithStreamingResponse",
+    "AsyncCheckpointsWithStreamingResponse",
+    "Jobs",
+    "AsyncJobs",
+    "JobsWithRawResponse",
+    "AsyncJobsWithRawResponse",
+    "JobsWithStreamingResponse",
+    "AsyncJobsWithStreamingResponse",
+]

backend/venv/lib/python3.12/site-packages/openai/resources/fine_tuning/jobs/checkpoints.py

@@ -0,0 +1,199 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+import httpx
+
+from .... import _legacy_response
+from ...._types import NOT_GIVEN, Body, Query, Headers, NotGiven
+from ...._utils import maybe_transform
+from ...._compat import cached_property
+from ...._resource import SyncAPIResource, AsyncAPIResource
+from ...._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
+from ....pagination import SyncCursorPage, AsyncCursorPage
+from ...._base_client import (
+    AsyncPaginator,
+    make_request_options,
+)
+from ....types.fine_tuning.jobs import checkpoint_list_params
+from ....types.fine_tuning.jobs.fine_tuning_job_checkpoint import FineTuningJobCheckpoint
+
+__all__ = ["Checkpoints", "AsyncCheckpoints"]
+
+
+class Checkpoints(SyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> CheckpointsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return CheckpointsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> CheckpointsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return CheckpointsWithStreamingResponse(self)
+
+    def list(
+        self,
+        fine_tuning_job_id: str,
+        *,
+        after: str | NotGiven = NOT_GIVEN,
+        limit: int | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> SyncCursorPage[FineTuningJobCheckpoint]:
+        """
+        List checkpoints for a fine-tuning job.
+
+        Args:
+          after: Identifier for the last checkpoint ID from the previous pagination request.
+
+          limit: Number of checkpoints to retrieve.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not fine_tuning_job_id:
+            raise ValueError(f"Expected a non-empty value for `fine_tuning_job_id` but received {fine_tuning_job_id!r}")
+        return self._get_api_list(
+            f"/fine_tuning/jobs/{fine_tuning_job_id}/checkpoints",
+            page=SyncCursorPage[FineTuningJobCheckpoint],
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform(
+                    {
+                        "after": after,
+                        "limit": limit,
+                    },
+                    checkpoint_list_params.CheckpointListParams,
+                ),
+            ),
+            model=FineTuningJobCheckpoint,
+        )
+
+
+class AsyncCheckpoints(AsyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> AsyncCheckpointsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncCheckpointsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncCheckpointsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncCheckpointsWithStreamingResponse(self)
+
+    def list(
+        self,
+        fine_tuning_job_id: str,
+        *,
+        after: str | NotGiven = NOT_GIVEN,
+        limit: int | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> AsyncPaginator[FineTuningJobCheckpoint, AsyncCursorPage[FineTuningJobCheckpoint]]:
+        """
+        List checkpoints for a fine-tuning job.
+
+        Args:
+          after: Identifier for the last checkpoint ID from the previous pagination request.
+
+          limit: Number of checkpoints to retrieve.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not fine_tuning_job_id:
+            raise ValueError(f"Expected a non-empty value for `fine_tuning_job_id` but received {fine_tuning_job_id!r}")
+        return self._get_api_list(
+            f"/fine_tuning/jobs/{fine_tuning_job_id}/checkpoints",
+            page=AsyncCursorPage[FineTuningJobCheckpoint],
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform(
+                    {
+                        "after": after,
+                        "limit": limit,
+                    },
+                    checkpoint_list_params.CheckpointListParams,
+                ),
+            ),
+            model=FineTuningJobCheckpoint,
+        )
+
+
+class CheckpointsWithRawResponse:
+    def __init__(self, checkpoints: Checkpoints) -> None:
+        self._checkpoints = checkpoints
+
+        self.list = _legacy_response.to_raw_response_wrapper(
+            checkpoints.list,
+        )
+
+
+class AsyncCheckpointsWithRawResponse:
+    def __init__(self, checkpoints: AsyncCheckpoints) -> None:
+        self._checkpoints = checkpoints
+
+        self.list = _legacy_response.async_to_raw_response_wrapper(
+            checkpoints.list,
+        )
+
+
+class CheckpointsWithStreamingResponse:
+    def __init__(self, checkpoints: Checkpoints) -> None:
+        self._checkpoints = checkpoints
+
+        self.list = to_streamed_response_wrapper(
+            checkpoints.list,
+        )
+
+
+class AsyncCheckpointsWithStreamingResponse:
+    def __init__(self, checkpoints: AsyncCheckpoints) -> None:
+        self._checkpoints = checkpoints
+
+        self.list = async_to_streamed_response_wrapper(
+            checkpoints.list,
+        )

backend/venv/lib/python3.12/site-packages/openai/resources/fine_tuning/jobs/jobs.py

@@ -0,0 +1,718 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Union, Iterable, Optional
+from typing_extensions import Literal
+
+import httpx
+
+from .... import _legacy_response
+from ...._types import NOT_GIVEN, Body, Query, Headers, NotGiven
+from ...._utils import (
+    maybe_transform,
+    async_maybe_transform,
+)
+from ...._compat import cached_property
+from .checkpoints import (
+    Checkpoints,
+    AsyncCheckpoints,
+    CheckpointsWithRawResponse,
+    AsyncCheckpointsWithRawResponse,
+    CheckpointsWithStreamingResponse,
+    AsyncCheckpointsWithStreamingResponse,
+)
+from ...._resource import SyncAPIResource, AsyncAPIResource
+from ...._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
+from ....pagination import SyncCursorPage, AsyncCursorPage
+from ...._base_client import (
+    AsyncPaginator,
+    make_request_options,
+)
+from ....types.fine_tuning import job_list_params, job_create_params, job_list_events_params
+from ....types.fine_tuning.fine_tuning_job import FineTuningJob
+from ....types.fine_tuning.fine_tuning_job_event import FineTuningJobEvent
+
+__all__ = ["Jobs", "AsyncJobs"]
+
+
+class Jobs(SyncAPIResource):
+    @cached_property
+    def checkpoints(self) -> Checkpoints:
+        return Checkpoints(self._client)
+
+    @cached_property
+    def with_raw_response(self) -> JobsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return JobsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> JobsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return JobsWithStreamingResponse(self)
+
+    def create(
+        self,
+        *,
+        model: Union[str, Literal["babbage-002", "davinci-002", "gpt-3.5-turbo", "gpt-4o-mini"]],
+        training_file: str,
+        hyperparameters: job_create_params.Hyperparameters | NotGiven = NOT_GIVEN,
+        integrations: Optional[Iterable[job_create_params.Integration]] | NotGiven = NOT_GIVEN,
+        seed: Optional[int] | NotGiven = NOT_GIVEN,
+        suffix: Optional[str] | NotGiven = NOT_GIVEN,
+        validation_file: Optional[str] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> FineTuningJob:
+        """
+        Creates a fine-tuning job which begins the process of creating a new model from
+        a given dataset.
+
+        Response includes details of the enqueued job including job status and the name
+        of the fine-tuned models once complete.
+
+        [Learn more about fine-tuning](https://platform.openai.com/docs/guides/fine-tuning)
+
+        Args:
+          model: The name of the model to fine-tune. You can select one of the
+              [supported models](https://platform.openai.com/docs/guides/fine-tuning/which-models-can-be-fine-tuned).
+
+          training_file: The ID of an uploaded file that contains training data.
+
+              See [upload file](https://platform.openai.com/docs/api-reference/files/create)
+              for how to upload a file.
+
+              Your dataset must be formatted as a JSONL file. Additionally, you must upload
+              your file with the purpose `fine-tune`.
+
+              The contents of the file should differ depending on if the model uses the
+              [chat](https://platform.openai.com/docs/api-reference/fine-tuning/chat-input) or
+              [completions](https://platform.openai.com/docs/api-reference/fine-tuning/completions-input)
+              format.
+
+              See the [fine-tuning guide](https://platform.openai.com/docs/guides/fine-tuning)
+              for more details.
+
+          hyperparameters: The hyperparameters used for the fine-tuning job.
+
+          integrations: A list of integrations to enable for your fine-tuning job.
+
+          seed: The seed controls the reproducibility of the job. Passing in the same seed and
+              job parameters should produce the same results, but may differ in rare cases. If
+              a seed is not specified, one will be generated for you.
+
+          suffix: A string of up to 64 characters that will be added to your fine-tuned model
+              name.
+
+              For example, a `suffix` of "custom-model-name" would produce a model name like
+              `ft:gpt-4o-mini:openai:custom-model-name:7p4lURel`.
+
+          validation_file: The ID of an uploaded file that contains validation data.
+
+              If you provide this file, the data is used to generate validation metrics
+              periodically during fine-tuning. These metrics can be viewed in the fine-tuning
+              results file. The same data should not be present in both train and validation
+              files.
+
+              Your dataset must be formatted as a JSONL file. You must upload your file with
+              the purpose `fine-tune`.
+
+              See the [fine-tuning guide](https://platform.openai.com/docs/guides/fine-tuning)
+              for more details.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        return self._post(
+            "/fine_tuning/jobs",
+            body=maybe_transform(
+                {
+                    "model": model,
+                    "training_file": training_file,
+                    "hyperparameters": hyperparameters,
+                    "integrations": integrations,
+                    "seed": seed,
+                    "suffix": suffix,
+                    "validation_file": validation_file,
+                },
+                job_create_params.JobCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=FineTuningJob,
+        )
+
+    def retrieve(
+        self,
+        fine_tuning_job_id: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> FineTuningJob:
+        """
+        Get info about a fine-tuning job.
+
+        [Learn more about fine-tuning](https://platform.openai.com/docs/guides/fine-tuning)
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not fine_tuning_job_id:
+            raise ValueError(f"Expected a non-empty value for `fine_tuning_job_id` but received {fine_tuning_job_id!r}")
+        return self._get(
+            f"/fine_tuning/jobs/{fine_tuning_job_id}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=FineTuningJob,
+        )
+
+    def list(
+        self,
+        *,
+        after: str | NotGiven = NOT_GIVEN,
+        limit: int | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> SyncCursorPage[FineTuningJob]:
+        """
+        List your organization's fine-tuning jobs
+
+        Args:
+          after: Identifier for the last job from the previous pagination request.
+
+          limit: Number of fine-tuning jobs to retrieve.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        return self._get_api_list(
+            "/fine_tuning/jobs",
+            page=SyncCursorPage[FineTuningJob],
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform(
+                    {
+                        "after": after,
+                        "limit": limit,
+                    },
+                    job_list_params.JobListParams,
+                ),
+            ),
+            model=FineTuningJob,
+        )
+
+    def cancel(
+        self,
+        fine_tuning_job_id: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> FineTuningJob:
+        """
+        Immediately cancel a fine-tune job.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not fine_tuning_job_id:
+            raise ValueError(f"Expected a non-empty value for `fine_tuning_job_id` but received {fine_tuning_job_id!r}")
+        return self._post(
+            f"/fine_tuning/jobs/{fine_tuning_job_id}/cancel",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=FineTuningJob,
+        )
+
+    def list_events(
+        self,
+        fine_tuning_job_id: str,
+        *,
+        after: str | NotGiven = NOT_GIVEN,
+        limit: int | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> SyncCursorPage[FineTuningJobEvent]:
+        """
+        Get status updates for a fine-tuning job.
+
+        Args:
+          after: Identifier for the last event from the previous pagination request.
+
+          limit: Number of events to retrieve.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not fine_tuning_job_id:
+            raise ValueError(f"Expected a non-empty value for `fine_tuning_job_id` but received {fine_tuning_job_id!r}")
+        return self._get_api_list(
+            f"/fine_tuning/jobs/{fine_tuning_job_id}/events",
+            page=SyncCursorPage[FineTuningJobEvent],
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform(
+                    {
+                        "after": after,
+                        "limit": limit,
+                    },
+                    job_list_events_params.JobListEventsParams,
+                ),
+            ),
+            model=FineTuningJobEvent,
+        )
+
+
+class AsyncJobs(AsyncAPIResource):
+    @cached_property
+    def checkpoints(self) -> AsyncCheckpoints:
+        return AsyncCheckpoints(self._client)
+
+    @cached_property
+    def with_raw_response(self) -> AsyncJobsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncJobsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncJobsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncJobsWithStreamingResponse(self)
+
+    async def create(
+        self,
+        *,
+        model: Union[str, Literal["babbage-002", "davinci-002", "gpt-3.5-turbo", "gpt-4o-mini"]],
+        training_file: str,
+        hyperparameters: job_create_params.Hyperparameters | NotGiven = NOT_GIVEN,
+        integrations: Optional[Iterable[job_create_params.Integration]] | NotGiven = NOT_GIVEN,
+        seed: Optional[int] | NotGiven = NOT_GIVEN,
+        suffix: Optional[str] | NotGiven = NOT_GIVEN,
+        validation_file: Optional[str] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> FineTuningJob:
+        """
+        Creates a fine-tuning job which begins the process of creating a new model from
+        a given dataset.
+
+        Response includes details of the enqueued job including job status and the name
+        of the fine-tuned models once complete.
+
+        [Learn more about fine-tuning](https://platform.openai.com/docs/guides/fine-tuning)
+
+        Args:
+          model: The name of the model to fine-tune. You can select one of the
+              [supported models](https://platform.openai.com/docs/guides/fine-tuning/which-models-can-be-fine-tuned).
+
+          training_file: The ID of an uploaded file that contains training data.
+
+              See [upload file](https://platform.openai.com/docs/api-reference/files/create)
+              for how to upload a file.
+
+              Your dataset must be formatted as a JSONL file. Additionally, you must upload
+              your file with the purpose `fine-tune`.
+
+              The contents of the file should differ depending on if the model uses the
+              [chat](https://platform.openai.com/docs/api-reference/fine-tuning/chat-input) or
+              [completions](https://platform.openai.com/docs/api-reference/fine-tuning/completions-input)
+              format.
+
+              See the [fine-tuning guide](https://platform.openai.com/docs/guides/fine-tuning)
+              for more details.
+
+          hyperparameters: The hyperparameters used for the fine-tuning job.
+
+          integrations: A list of integrations to enable for your fine-tuning job.
+
+          seed: The seed controls the reproducibility of the job. Passing in the same seed and
+              job parameters should produce the same results, but may differ in rare cases. If
+              a seed is not specified, one will be generated for you.
+
+          suffix: A string of up to 64 characters that will be added to your fine-tuned model
+              name.
+
+              For example, a `suffix` of "custom-model-name" would produce a model name like
+              `ft:gpt-4o-mini:openai:custom-model-name:7p4lURel`.
+
+          validation_file: The ID of an uploaded file that contains validation data.
+
+              If you provide this file, the data is used to generate validation metrics
+              periodically during fine-tuning. These metrics can be viewed in the fine-tuning
+              results file. The same data should not be present in both train and validation
+              files.
+
+              Your dataset must be formatted as a JSONL file. You must upload your file with
+              the purpose `fine-tune`.
+
+              See the [fine-tuning guide](https://platform.openai.com/docs/guides/fine-tuning)
+              for more details.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        return await self._post(
+            "/fine_tuning/jobs",
+            body=await async_maybe_transform(
+                {
+                    "model": model,
+                    "training_file": training_file,
+                    "hyperparameters": hyperparameters,
+                    "integrations": integrations,
+                    "seed": seed,
+                    "suffix": suffix,
+                    "validation_file": validation_file,
+                },
+                job_create_params.JobCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=FineTuningJob,
+        )
+
+    async def retrieve(
+        self,
+        fine_tuning_job_id: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> FineTuningJob:
+        """
+        Get info about a fine-tuning job.
+
+        [Learn more about fine-tuning](https://platform.openai.com/docs/guides/fine-tuning)
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not fine_tuning_job_id:
+            raise ValueError(f"Expected a non-empty value for `fine_tuning_job_id` but received {fine_tuning_job_id!r}")
+        return await self._get(
+            f"/fine_tuning/jobs/{fine_tuning_job_id}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=FineTuningJob,
+        )
+
+    def list(
+        self,
+        *,
+        after: str | NotGiven = NOT_GIVEN,
+        limit: int | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> AsyncPaginator[FineTuningJob, AsyncCursorPage[FineTuningJob]]:
+        """
+        List your organization's fine-tuning jobs
+
+        Args:
+          after: Identifier for the last job from the previous pagination request.
+
+          limit: Number of fine-tuning jobs to retrieve.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        return self._get_api_list(
+            "/fine_tuning/jobs",
+            page=AsyncCursorPage[FineTuningJob],
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform(
+                    {
+                        "after": after,
+                        "limit": limit,
+                    },
+                    job_list_params.JobListParams,
+                ),
+            ),
+            model=FineTuningJob,
+        )
+
+    async def cancel(
+        self,
+        fine_tuning_job_id: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> FineTuningJob:
+        """
+        Immediately cancel a fine-tune job.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not fine_tuning_job_id:
+            raise ValueError(f"Expected a non-empty value for `fine_tuning_job_id` but received {fine_tuning_job_id!r}")
+        return await self._post(
+            f"/fine_tuning/jobs/{fine_tuning_job_id}/cancel",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=FineTuningJob,
+        )
+
+    def list_events(
+        self,
+        fine_tuning_job_id: str,
+        *,
+        after: str | NotGiven = NOT_GIVEN,
+        limit: int | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> AsyncPaginator[FineTuningJobEvent, AsyncCursorPage[FineTuningJobEvent]]:
+        """
+        Get status updates for a fine-tuning job.
+
+        Args:
+          after: Identifier for the last event from the previous pagination request.
+
+          limit: Number of events to retrieve.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not fine_tuning_job_id:
+            raise ValueError(f"Expected a non-empty value for `fine_tuning_job_id` but received {fine_tuning_job_id!r}")
+        return self._get_api_list(
+            f"/fine_tuning/jobs/{fine_tuning_job_id}/events",
+            page=AsyncCursorPage[FineTuningJobEvent],
+            options=make_request_options(
+                extra_headers=extra_headers,
+                extra_query=extra_query,
+                extra_body=extra_body,
+                timeout=timeout,
+                query=maybe_transform(
+                    {
+                        "after": after,
+                        "limit": limit,
+                    },
+                    job_list_events_params.JobListEventsParams,
+                ),
+            ),
+            model=FineTuningJobEvent,
+        )
+
+
+class JobsWithRawResponse:
+    def __init__(self, jobs: Jobs) -> None:
+        self._jobs = jobs
+
+        self.create = _legacy_response.to_raw_response_wrapper(
+            jobs.create,
+        )
+        self.retrieve = _legacy_response.to_raw_response_wrapper(
+            jobs.retrieve,
+        )
+        self.list = _legacy_response.to_raw_response_wrapper(
+            jobs.list,
+        )
+        self.cancel = _legacy_response.to_raw_response_wrapper(
+            jobs.cancel,
+        )
+        self.list_events = _legacy_response.to_raw_response_wrapper(
+            jobs.list_events,
+        )
+
+    @cached_property
+    def checkpoints(self) -> CheckpointsWithRawResponse:
+        return CheckpointsWithRawResponse(self._jobs.checkpoints)
+
+
+class AsyncJobsWithRawResponse:
+    def __init__(self, jobs: AsyncJobs) -> None:
+        self._jobs = jobs
+
+        self.create = _legacy_response.async_to_raw_response_wrapper(
+            jobs.create,
+        )
+        self.retrieve = _legacy_response.async_to_raw_response_wrapper(
+            jobs.retrieve,
+        )
+        self.list = _legacy_response.async_to_raw_response_wrapper(
+            jobs.list,
+        )
+        self.cancel = _legacy_response.async_to_raw_response_wrapper(
+            jobs.cancel,
+        )
+        self.list_events = _legacy_response.async_to_raw_response_wrapper(
+            jobs.list_events,
+        )
+
+    @cached_property
+    def checkpoints(self) -> AsyncCheckpointsWithRawResponse:
+        return AsyncCheckpointsWithRawResponse(self._jobs.checkpoints)
+
+
+class JobsWithStreamingResponse:
+    def __init__(self, jobs: Jobs) -> None:
+        self._jobs = jobs
+
+        self.create = to_streamed_response_wrapper(
+            jobs.create,
+        )
+        self.retrieve = to_streamed_response_wrapper(
+            jobs.retrieve,
+        )
+        self.list = to_streamed_response_wrapper(
+            jobs.list,
+        )
+        self.cancel = to_streamed_response_wrapper(
+            jobs.cancel,
+        )
+        self.list_events = to_streamed_response_wrapper(
+            jobs.list_events,
+        )
+
+    @cached_property
+    def checkpoints(self) -> CheckpointsWithStreamingResponse:
+        return CheckpointsWithStreamingResponse(self._jobs.checkpoints)
+
+
+class AsyncJobsWithStreamingResponse:
+    def __init__(self, jobs: AsyncJobs) -> None:
+        self._jobs = jobs
+
+        self.create = async_to_streamed_response_wrapper(
+            jobs.create,
+        )
+        self.retrieve = async_to_streamed_response_wrapper(
+            jobs.retrieve,
+        )
+        self.list = async_to_streamed_response_wrapper(
+            jobs.list,
+        )
+        self.cancel = async_to_streamed_response_wrapper(
+            jobs.cancel,
+        )
+        self.list_events = async_to_streamed_response_wrapper(
+            jobs.list_events,
+        )
+
+    @cached_property
+    def checkpoints(self) -> AsyncCheckpointsWithStreamingResponse:
+        return AsyncCheckpointsWithStreamingResponse(self._jobs.checkpoints)

backend/venv/lib/python3.12/site-packages/openai/resources/images.py

@@ -0,0 +1,600 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Union, Mapping, Optional, cast
+from typing_extensions import Literal
+
+import httpx
+
+from .. import _legacy_response
+from ..types import image_edit_params, image_generate_params, image_create_variation_params
+from .._types import NOT_GIVEN, Body, Query, Headers, NotGiven, FileTypes
+from .._utils import (
+    extract_files,
+    maybe_transform,
+    deepcopy_minimal,
+    async_maybe_transform,
+)
+from .._compat import cached_property
+from .._resource import SyncAPIResource, AsyncAPIResource
+from .._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
+from .._base_client import make_request_options
+from ..types.image_model import ImageModel
+from ..types.images_response import ImagesResponse
+
+__all__ = ["Images", "AsyncImages"]
+
+
+class Images(SyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> ImagesWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return ImagesWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> ImagesWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return ImagesWithStreamingResponse(self)
+
+    def create_variation(
+        self,
+        *,
+        image: FileTypes,
+        model: Union[str, ImageModel, None] | NotGiven = NOT_GIVEN,
+        n: Optional[int] | NotGiven = NOT_GIVEN,
+        response_format: Optional[Literal["url", "b64_json"]] | NotGiven = NOT_GIVEN,
+        size: Optional[Literal["256x256", "512x512", "1024x1024"]] | NotGiven = NOT_GIVEN,
+        user: str | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> ImagesResponse:
+        """
+        Creates a variation of a given image.
+
+        Args:
+          image: The image to use as the basis for the variation(s). Must be a valid PNG file,
+              less than 4MB, and square.
+
+          model: The model to use for image generation. Only `dall-e-2` is supported at this
+              time.
+
+          n: The number of images to generate. Must be between 1 and 10. For `dall-e-3`, only
+              `n=1` is supported.
+
+          response_format: The format in which the generated images are returned. Must be one of `url` or
+              `b64_json`. URLs are only valid for 60 minutes after the image has been
+              generated.
+
+          size: The size of the generated images. Must be one of `256x256`, `512x512`, or
+              `1024x1024`.
+
+          user: A unique identifier representing your end-user, which can help OpenAI to monitor
+              and detect abuse.
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        body = deepcopy_minimal(
+            {
+                "image": image,
+                "model": model,
+                "n": n,
+                "response_format": response_format,
+                "size": size,
+                "user": user,
+            }
+        )
+        files = extract_files(cast(Mapping[str, object], body), paths=[["image"]])
+        # It should be noted that the actual Content-Type header that will be
+        # sent to the server will contain a `boundary` parameter, e.g.
+        # multipart/form-data; boundary=---abc--
+        extra_headers = {"Content-Type": "multipart/form-data", **(extra_headers or {})}
+        return self._post(
+            "/images/variations",
+            body=maybe_transform(body, image_create_variation_params.ImageCreateVariationParams),
+            files=files,
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=ImagesResponse,
+        )
+
+    def edit(
+        self,
+        *,
+        image: FileTypes,
+        prompt: str,
+        mask: FileTypes | NotGiven = NOT_GIVEN,
+        model: Union[str, ImageModel, None] | NotGiven = NOT_GIVEN,
+        n: Optional[int] | NotGiven = NOT_GIVEN,
+        response_format: Optional[Literal["url", "b64_json"]] | NotGiven = NOT_GIVEN,
+        size: Optional[Literal["256x256", "512x512", "1024x1024"]] | NotGiven = NOT_GIVEN,
+        user: str | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> ImagesResponse:
+        """
+        Creates an edited or extended image given an original image and a prompt.
+
+        Args:
+          image: The image to edit. Must be a valid PNG file, less than 4MB, and square. If mask
+              is not provided, image must have transparency, which will be used as the mask.
+
+          prompt: A text description of the desired image(s). The maximum length is 1000
+              characters.
+
+          mask: An additional image whose fully transparent areas (e.g. where alpha is zero)
+              indicate where `image` should be edited. Must be a valid PNG file, less than
+              4MB, and have the same dimensions as `image`.
+
+          model: The model to use for image generation. Only `dall-e-2` is supported at this
+              time.
+
+          n: The number of images to generate. Must be between 1 and 10.
+
+          response_format: The format in which the generated images are returned. Must be one of `url` or
+              `b64_json`. URLs are only valid for 60 minutes after the image has been
+              generated.
+
+          size: The size of the generated images. Must be one of `256x256`, `512x512`, or
+              `1024x1024`.
+
+          user: A unique identifier representing your end-user, which can help OpenAI to monitor
+              and detect abuse.
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        body = deepcopy_minimal(
+            {
+                "image": image,
+                "prompt": prompt,
+                "mask": mask,
+                "model": model,
+                "n": n,
+                "response_format": response_format,
+                "size": size,
+                "user": user,
+            }
+        )
+        files = extract_files(cast(Mapping[str, object], body), paths=[["image"], ["mask"]])
+        # It should be noted that the actual Content-Type header that will be
+        # sent to the server will contain a `boundary` parameter, e.g.
+        # multipart/form-data; boundary=---abc--
+        extra_headers = {"Content-Type": "multipart/form-data", **(extra_headers or {})}
+        return self._post(
+            "/images/edits",
+            body=maybe_transform(body, image_edit_params.ImageEditParams),
+            files=files,
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=ImagesResponse,
+        )
+
+    def generate(
+        self,
+        *,
+        prompt: str,
+        model: Union[str, ImageModel, None] | NotGiven = NOT_GIVEN,
+        n: Optional[int] | NotGiven = NOT_GIVEN,
+        quality: Literal["standard", "hd"] | NotGiven = NOT_GIVEN,
+        response_format: Optional[Literal["url", "b64_json"]] | NotGiven = NOT_GIVEN,
+        size: Optional[Literal["256x256", "512x512", "1024x1024", "1792x1024", "1024x1792"]] | NotGiven = NOT_GIVEN,
+        style: Optional[Literal["vivid", "natural"]] | NotGiven = NOT_GIVEN,
+        user: str | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> ImagesResponse:
+        """
+        Creates an image given a prompt.
+
+        Args:
+          prompt: A text description of the desired image(s). The maximum length is 1000
+              characters for `dall-e-2` and 4000 characters for `dall-e-3`.
+
+          model: The model to use for image generation.
+
+          n: The number of images to generate. Must be between 1 and 10. For `dall-e-3`, only
+              `n=1` is supported.
+
+          quality: The quality of the image that will be generated. `hd` creates images with finer
+              details and greater consistency across the image. This param is only supported
+              for `dall-e-3`.
+
+          response_format: The format in which the generated images are returned. Must be one of `url` or
+              `b64_json`. URLs are only valid for 60 minutes after the image has been
+              generated.
+
+          size: The size of the generated images. Must be one of `256x256`, `512x512`, or
+              `1024x1024` for `dall-e-2`. Must be one of `1024x1024`, `1792x1024`, or
+              `1024x1792` for `dall-e-3` models.
+
+          style: The style of the generated images. Must be one of `vivid` or `natural`. Vivid
+              causes the model to lean towards generating hyper-real and dramatic images.
+              Natural causes the model to produce more natural, less hyper-real looking
+              images. This param is only supported for `dall-e-3`.
+
+          user: A unique identifier representing your end-user, which can help OpenAI to monitor
+              and detect abuse.
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        return self._post(
+            "/images/generations",
+            body=maybe_transform(
+                {
+                    "prompt": prompt,
+                    "model": model,
+                    "n": n,
+                    "quality": quality,
+                    "response_format": response_format,
+                    "size": size,
+                    "style": style,
+                    "user": user,
+                },
+                image_generate_params.ImageGenerateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=ImagesResponse,
+        )
+
+
+class AsyncImages(AsyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> AsyncImagesWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncImagesWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncImagesWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncImagesWithStreamingResponse(self)
+
+    async def create_variation(
+        self,
+        *,
+        image: FileTypes,
+        model: Union[str, ImageModel, None] | NotGiven = NOT_GIVEN,
+        n: Optional[int] | NotGiven = NOT_GIVEN,
+        response_format: Optional[Literal["url", "b64_json"]] | NotGiven = NOT_GIVEN,
+        size: Optional[Literal["256x256", "512x512", "1024x1024"]] | NotGiven = NOT_GIVEN,
+        user: str | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> ImagesResponse:
+        """
+        Creates a variation of a given image.
+
+        Args:
+          image: The image to use as the basis for the variation(s). Must be a valid PNG file,
+              less than 4MB, and square.
+
+          model: The model to use for image generation. Only `dall-e-2` is supported at this
+              time.
+
+          n: The number of images to generate. Must be between 1 and 10. For `dall-e-3`, only
+              `n=1` is supported.
+
+          response_format: The format in which the generated images are returned. Must be one of `url` or
+              `b64_json`. URLs are only valid for 60 minutes after the image has been
+              generated.
+
+          size: The size of the generated images. Must be one of `256x256`, `512x512`, or
+              `1024x1024`.
+
+          user: A unique identifier representing your end-user, which can help OpenAI to monitor
+              and detect abuse.
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        body = deepcopy_minimal(
+            {
+                "image": image,
+                "model": model,
+                "n": n,
+                "response_format": response_format,
+                "size": size,
+                "user": user,
+            }
+        )
+        files = extract_files(cast(Mapping[str, object], body), paths=[["image"]])
+        # It should be noted that the actual Content-Type header that will be
+        # sent to the server will contain a `boundary` parameter, e.g.
+        # multipart/form-data; boundary=---abc--
+        extra_headers = {"Content-Type": "multipart/form-data", **(extra_headers or {})}
+        return await self._post(
+            "/images/variations",
+            body=await async_maybe_transform(body, image_create_variation_params.ImageCreateVariationParams),
+            files=files,
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=ImagesResponse,
+        )
+
+    async def edit(
+        self,
+        *,
+        image: FileTypes,
+        prompt: str,
+        mask: FileTypes | NotGiven = NOT_GIVEN,
+        model: Union[str, ImageModel, None] | NotGiven = NOT_GIVEN,
+        n: Optional[int] | NotGiven = NOT_GIVEN,
+        response_format: Optional[Literal["url", "b64_json"]] | NotGiven = NOT_GIVEN,
+        size: Optional[Literal["256x256", "512x512", "1024x1024"]] | NotGiven = NOT_GIVEN,
+        user: str | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> ImagesResponse:
+        """
+        Creates an edited or extended image given an original image and a prompt.
+
+        Args:
+          image: The image to edit. Must be a valid PNG file, less than 4MB, and square. If mask
+              is not provided, image must have transparency, which will be used as the mask.
+
+          prompt: A text description of the desired image(s). The maximum length is 1000
+              characters.
+
+          mask: An additional image whose fully transparent areas (e.g. where alpha is zero)
+              indicate where `image` should be edited. Must be a valid PNG file, less than
+              4MB, and have the same dimensions as `image`.
+
+          model: The model to use for image generation. Only `dall-e-2` is supported at this
+              time.
+
+          n: The number of images to generate. Must be between 1 and 10.
+
+          response_format: The format in which the generated images are returned. Must be one of `url` or
+              `b64_json`. URLs are only valid for 60 minutes after the image has been
+              generated.
+
+          size: The size of the generated images. Must be one of `256x256`, `512x512`, or
+              `1024x1024`.
+
+          user: A unique identifier representing your end-user, which can help OpenAI to monitor
+              and detect abuse.
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        body = deepcopy_minimal(
+            {
+                "image": image,
+                "prompt": prompt,
+                "mask": mask,
+                "model": model,
+                "n": n,
+                "response_format": response_format,
+                "size": size,
+                "user": user,
+            }
+        )
+        files = extract_files(cast(Mapping[str, object], body), paths=[["image"], ["mask"]])
+        # It should be noted that the actual Content-Type header that will be
+        # sent to the server will contain a `boundary` parameter, e.g.
+        # multipart/form-data; boundary=---abc--
+        extra_headers = {"Content-Type": "multipart/form-data", **(extra_headers or {})}
+        return await self._post(
+            "/images/edits",
+            body=await async_maybe_transform(body, image_edit_params.ImageEditParams),
+            files=files,
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=ImagesResponse,
+        )
+
+    async def generate(
+        self,
+        *,
+        prompt: str,
+        model: Union[str, ImageModel, None] | NotGiven = NOT_GIVEN,
+        n: Optional[int] | NotGiven = NOT_GIVEN,
+        quality: Literal["standard", "hd"] | NotGiven = NOT_GIVEN,
+        response_format: Optional[Literal["url", "b64_json"]] | NotGiven = NOT_GIVEN,
+        size: Optional[Literal["256x256", "512x512", "1024x1024", "1792x1024", "1024x1792"]] | NotGiven = NOT_GIVEN,
+        style: Optional[Literal["vivid", "natural"]] | NotGiven = NOT_GIVEN,
+        user: str | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> ImagesResponse:
+        """
+        Creates an image given a prompt.
+
+        Args:
+          prompt: A text description of the desired image(s). The maximum length is 1000
+              characters for `dall-e-2` and 4000 characters for `dall-e-3`.
+
+          model: The model to use for image generation.
+
+          n: The number of images to generate. Must be between 1 and 10. For `dall-e-3`, only
+              `n=1` is supported.
+
+          quality: The quality of the image that will be generated. `hd` creates images with finer
+              details and greater consistency across the image. This param is only supported
+              for `dall-e-3`.
+
+          response_format: The format in which the generated images are returned. Must be one of `url` or
+              `b64_json`. URLs are only valid for 60 minutes after the image has been
+              generated.
+
+          size: The size of the generated images. Must be one of `256x256`, `512x512`, or
+              `1024x1024` for `dall-e-2`. Must be one of `1024x1024`, `1792x1024`, or
+              `1024x1792` for `dall-e-3` models.
+
+          style: The style of the generated images. Must be one of `vivid` or `natural`. Vivid
+              causes the model to lean towards generating hyper-real and dramatic images.
+              Natural causes the model to produce more natural, less hyper-real looking
+              images. This param is only supported for `dall-e-3`.
+
+          user: A unique identifier representing your end-user, which can help OpenAI to monitor
+              and detect abuse.
+              [Learn more](https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids).
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        return await self._post(
+            "/images/generations",
+            body=await async_maybe_transform(
+                {
+                    "prompt": prompt,
+                    "model": model,
+                    "n": n,
+                    "quality": quality,
+                    "response_format": response_format,
+                    "size": size,
+                    "style": style,
+                    "user": user,
+                },
+                image_generate_params.ImageGenerateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=ImagesResponse,
+        )
+
+
+class ImagesWithRawResponse:
+    def __init__(self, images: Images) -> None:
+        self._images = images
+
+        self.create_variation = _legacy_response.to_raw_response_wrapper(
+            images.create_variation,
+        )
+        self.edit = _legacy_response.to_raw_response_wrapper(
+            images.edit,
+        )
+        self.generate = _legacy_response.to_raw_response_wrapper(
+            images.generate,
+        )
+
+
+class AsyncImagesWithRawResponse:
+    def __init__(self, images: AsyncImages) -> None:
+        self._images = images
+
+        self.create_variation = _legacy_response.async_to_raw_response_wrapper(
+            images.create_variation,
+        )
+        self.edit = _legacy_response.async_to_raw_response_wrapper(
+            images.edit,
+        )
+        self.generate = _legacy_response.async_to_raw_response_wrapper(
+            images.generate,
+        )
+
+
+class ImagesWithStreamingResponse:
+    def __init__(self, images: Images) -> None:
+        self._images = images
+
+        self.create_variation = to_streamed_response_wrapper(
+            images.create_variation,
+        )
+        self.edit = to_streamed_response_wrapper(
+            images.edit,
+        )
+        self.generate = to_streamed_response_wrapper(
+            images.generate,
+        )
+
+
+class AsyncImagesWithStreamingResponse:
+    def __init__(self, images: AsyncImages) -> None:
+        self._images = images
+
+        self.create_variation = async_to_streamed_response_wrapper(
+            images.create_variation,
+        )
+        self.edit = async_to_streamed_response_wrapper(
+            images.edit,
+        )
+        self.generate = async_to_streamed_response_wrapper(
+            images.generate,
+        )

backend/venv/lib/python3.12/site-packages/openai/resources/models.py

@@ -0,0 +1,306 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+import httpx
+
+from .. import _legacy_response
+from .._types import NOT_GIVEN, Body, Query, Headers, NotGiven
+from .._compat import cached_property
+from .._resource import SyncAPIResource, AsyncAPIResource
+from .._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
+from ..pagination import SyncPage, AsyncPage
+from ..types.model import Model
+from .._base_client import (
+    AsyncPaginator,
+    make_request_options,
+)
+from ..types.model_deleted import ModelDeleted
+
+__all__ = ["Models", "AsyncModels"]
+
+
+class Models(SyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> ModelsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return ModelsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> ModelsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return ModelsWithStreamingResponse(self)
+
+    def retrieve(
+        self,
+        model: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Model:
+        """
+        Retrieves a model instance, providing basic information about the model such as
+        the owner and permissioning.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not model:
+            raise ValueError(f"Expected a non-empty value for `model` but received {model!r}")
+        return self._get(
+            f"/models/{model}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=Model,
+        )
+
+    def list(
+        self,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> SyncPage[Model]:
+        """
+        Lists the currently available models, and provides basic information about each
+        one such as the owner and availability.
+        """
+        return self._get_api_list(
+            "/models",
+            page=SyncPage[Model],
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            model=Model,
+        )
+
+    def delete(
+        self,
+        model: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> ModelDeleted:
+        """Delete a fine-tuned model.
+
+        You must have the Owner role in your organization to
+        delete a model.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not model:
+            raise ValueError(f"Expected a non-empty value for `model` but received {model!r}")
+        return self._delete(
+            f"/models/{model}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=ModelDeleted,
+        )
+
+
+class AsyncModels(AsyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> AsyncModelsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncModelsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncModelsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncModelsWithStreamingResponse(self)
+
+    async def retrieve(
+        self,
+        model: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Model:
+        """
+        Retrieves a model instance, providing basic information about the model such as
+        the owner and permissioning.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not model:
+            raise ValueError(f"Expected a non-empty value for `model` but received {model!r}")
+        return await self._get(
+            f"/models/{model}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=Model,
+        )
+
+    def list(
+        self,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> AsyncPaginator[Model, AsyncPage[Model]]:
+        """
+        Lists the currently available models, and provides basic information about each
+        one such as the owner and availability.
+        """
+        return self._get_api_list(
+            "/models",
+            page=AsyncPage[Model],
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            model=Model,
+        )
+
+    async def delete(
+        self,
+        model: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> ModelDeleted:
+        """Delete a fine-tuned model.
+
+        You must have the Owner role in your organization to
+        delete a model.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not model:
+            raise ValueError(f"Expected a non-empty value for `model` but received {model!r}")
+        return await self._delete(
+            f"/models/{model}",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=ModelDeleted,
+        )
+
+
+class ModelsWithRawResponse:
+    def __init__(self, models: Models) -> None:
+        self._models = models
+
+        self.retrieve = _legacy_response.to_raw_response_wrapper(
+            models.retrieve,
+        )
+        self.list = _legacy_response.to_raw_response_wrapper(
+            models.list,
+        )
+        self.delete = _legacy_response.to_raw_response_wrapper(
+            models.delete,
+        )
+
+
+class AsyncModelsWithRawResponse:
+    def __init__(self, models: AsyncModels) -> None:
+        self._models = models
+
+        self.retrieve = _legacy_response.async_to_raw_response_wrapper(
+            models.retrieve,
+        )
+        self.list = _legacy_response.async_to_raw_response_wrapper(
+            models.list,
+        )
+        self.delete = _legacy_response.async_to_raw_response_wrapper(
+            models.delete,
+        )
+
+
+class ModelsWithStreamingResponse:
+    def __init__(self, models: Models) -> None:
+        self._models = models
+
+        self.retrieve = to_streamed_response_wrapper(
+            models.retrieve,
+        )
+        self.list = to_streamed_response_wrapper(
+            models.list,
+        )
+        self.delete = to_streamed_response_wrapper(
+            models.delete,
+        )
+
+
+class AsyncModelsWithStreamingResponse:
+    def __init__(self, models: AsyncModels) -> None:
+        self._models = models
+
+        self.retrieve = async_to_streamed_response_wrapper(
+            models.retrieve,
+        )
+        self.list = async_to_streamed_response_wrapper(
+            models.list,
+        )
+        self.delete = async_to_streamed_response_wrapper(
+            models.delete,
+        )

backend/venv/lib/python3.12/site-packages/openai/resources/moderations.py

@@ -0,0 +1,200 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import List, Union, Iterable
+
+import httpx
+
+from .. import _legacy_response
+from ..types import moderation_create_params
+from .._types import NOT_GIVEN, Body, Query, Headers, NotGiven
+from .._utils import (
+    maybe_transform,
+    async_maybe_transform,
+)
+from .._compat import cached_property
+from .._resource import SyncAPIResource, AsyncAPIResource
+from .._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
+from .._base_client import make_request_options
+from ..types.moderation_model import ModerationModel
+from ..types.moderation_create_response import ModerationCreateResponse
+from ..types.moderation_multi_modal_input_param import ModerationMultiModalInputParam
+
+__all__ = ["Moderations", "AsyncModerations"]
+
+
+class Moderations(SyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> ModerationsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return ModerationsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> ModerationsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return ModerationsWithStreamingResponse(self)
+
+    def create(
+        self,
+        *,
+        input: Union[str, List[str], Iterable[ModerationMultiModalInputParam]],
+        model: Union[str, ModerationModel] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> ModerationCreateResponse:
+        """Classifies if text and/or image inputs are potentially harmful.
+
+        Learn more in
+        the [moderation guide](https://platform.openai.com/docs/guides/moderation).
+
+        Args:
+          input: Input (or inputs) to classify. Can be a single string, an array of strings, or
+              an array of multi-modal input objects similar to other models.
+
+          model: The content moderation model you would like to use. Learn more in
+              [the moderation guide](https://platform.openai.com/docs/guides/moderation), and
+              learn about available models
+              [here](https://platform.openai.com/docs/models/moderation).
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        return self._post(
+            "/moderations",
+            body=maybe_transform(
+                {
+                    "input": input,
+                    "model": model,
+                },
+                moderation_create_params.ModerationCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=ModerationCreateResponse,
+        )
+
+
+class AsyncModerations(AsyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> AsyncModerationsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncModerationsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncModerationsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncModerationsWithStreamingResponse(self)
+
+    async def create(
+        self,
+        *,
+        input: Union[str, List[str], Iterable[ModerationMultiModalInputParam]],
+        model: Union[str, ModerationModel] | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> ModerationCreateResponse:
+        """Classifies if text and/or image inputs are potentially harmful.
+
+        Learn more in
+        the [moderation guide](https://platform.openai.com/docs/guides/moderation).
+
+        Args:
+          input: Input (or inputs) to classify. Can be a single string, an array of strings, or
+              an array of multi-modal input objects similar to other models.
+
+          model: The content moderation model you would like to use. Learn more in
+              [the moderation guide](https://platform.openai.com/docs/guides/moderation), and
+              learn about available models
+              [here](https://platform.openai.com/docs/models/moderation).
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        return await self._post(
+            "/moderations",
+            body=await async_maybe_transform(
+                {
+                    "input": input,
+                    "model": model,
+                },
+                moderation_create_params.ModerationCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=ModerationCreateResponse,
+        )
+
+
+class ModerationsWithRawResponse:
+    def __init__(self, moderations: Moderations) -> None:
+        self._moderations = moderations
+
+        self.create = _legacy_response.to_raw_response_wrapper(
+            moderations.create,
+        )
+
+
+class AsyncModerationsWithRawResponse:
+    def __init__(self, moderations: AsyncModerations) -> None:
+        self._moderations = moderations
+
+        self.create = _legacy_response.async_to_raw_response_wrapper(
+            moderations.create,
+        )
+
+
+class ModerationsWithStreamingResponse:
+    def __init__(self, moderations: Moderations) -> None:
+        self._moderations = moderations
+
+        self.create = to_streamed_response_wrapper(
+            moderations.create,
+        )
+
+
+class AsyncModerationsWithStreamingResponse:
+    def __init__(self, moderations: AsyncModerations) -> None:
+        self._moderations = moderations
+
+        self.create = async_to_streamed_response_wrapper(
+            moderations.create,
+        )

backend/venv/lib/python3.12/site-packages/openai/resources/uploads/__init__.py

@@ -0,0 +1,33 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from .parts import (
+    Parts,
+    AsyncParts,
+    PartsWithRawResponse,
+    AsyncPartsWithRawResponse,
+    PartsWithStreamingResponse,
+    AsyncPartsWithStreamingResponse,
+)
+from .uploads import (
+    Uploads,
+    AsyncUploads,
+    UploadsWithRawResponse,
+    AsyncUploadsWithRawResponse,
+    UploadsWithStreamingResponse,
+    AsyncUploadsWithStreamingResponse,
+)
+
+__all__ = [
+    "Parts",
+    "AsyncParts",
+    "PartsWithRawResponse",
+    "AsyncPartsWithRawResponse",
+    "PartsWithStreamingResponse",
+    "AsyncPartsWithStreamingResponse",
+    "Uploads",
+    "AsyncUploads",
+    "UploadsWithRawResponse",
+    "AsyncUploadsWithRawResponse",
+    "UploadsWithStreamingResponse",
+    "AsyncUploadsWithStreamingResponse",
+]

backend/venv/lib/python3.12/site-packages/openai/resources/uploads/parts.py

@@ -0,0 +1,210 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Mapping, cast
+
+import httpx
+
+from ... import _legacy_response
+from ..._types import NOT_GIVEN, Body, Query, Headers, NotGiven, FileTypes
+from ..._utils import (
+    extract_files,
+    maybe_transform,
+    deepcopy_minimal,
+    async_maybe_transform,
+)
+from ..._compat import cached_property
+from ..._resource import SyncAPIResource, AsyncAPIResource
+from ..._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
+from ..._base_client import make_request_options
+from ...types.uploads import part_create_params
+from ...types.uploads.upload_part import UploadPart
+
+__all__ = ["Parts", "AsyncParts"]
+
+
+class Parts(SyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> PartsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return PartsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> PartsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return PartsWithStreamingResponse(self)
+
+    def create(
+        self,
+        upload_id: str,
+        *,
+        data: FileTypes,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> UploadPart:
+        """
+        Adds a
+        [Part](https://platform.openai.com/docs/api-reference/uploads/part-object) to an
+        [Upload](https://platform.openai.com/docs/api-reference/uploads/object) object.
+        A Part represents a chunk of bytes from the file you are trying to upload.
+
+        Each Part can be at most 64 MB, and you can add Parts until you hit the Upload
+        maximum of 8 GB.
+
+        It is possible to add multiple Parts in parallel. You can decide the intended
+        order of the Parts when you
+        [complete the Upload](https://platform.openai.com/docs/api-reference/uploads/complete).
+
+        Args:
+          data: The chunk of bytes for this Part.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not upload_id:
+            raise ValueError(f"Expected a non-empty value for `upload_id` but received {upload_id!r}")
+        body = deepcopy_minimal({"data": data})
+        files = extract_files(cast(Mapping[str, object], body), paths=[["data"]])
+        # It should be noted that the actual Content-Type header that will be
+        # sent to the server will contain a `boundary` parameter, e.g.
+        # multipart/form-data; boundary=---abc--
+        extra_headers = {"Content-Type": "multipart/form-data", **(extra_headers or {})}
+        return self._post(
+            f"/uploads/{upload_id}/parts",
+            body=maybe_transform(body, part_create_params.PartCreateParams),
+            files=files,
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=UploadPart,
+        )
+
+
+class AsyncParts(AsyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> AsyncPartsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncPartsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncPartsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncPartsWithStreamingResponse(self)
+
+    async def create(
+        self,
+        upload_id: str,
+        *,
+        data: FileTypes,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> UploadPart:
+        """
+        Adds a
+        [Part](https://platform.openai.com/docs/api-reference/uploads/part-object) to an
+        [Upload](https://platform.openai.com/docs/api-reference/uploads/object) object.
+        A Part represents a chunk of bytes from the file you are trying to upload.
+
+        Each Part can be at most 64 MB, and you can add Parts until you hit the Upload
+        maximum of 8 GB.
+
+        It is possible to add multiple Parts in parallel. You can decide the intended
+        order of the Parts when you
+        [complete the Upload](https://platform.openai.com/docs/api-reference/uploads/complete).
+
+        Args:
+          data: The chunk of bytes for this Part.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not upload_id:
+            raise ValueError(f"Expected a non-empty value for `upload_id` but received {upload_id!r}")
+        body = deepcopy_minimal({"data": data})
+        files = extract_files(cast(Mapping[str, object], body), paths=[["data"]])
+        # It should be noted that the actual Content-Type header that will be
+        # sent to the server will contain a `boundary` parameter, e.g.
+        # multipart/form-data; boundary=---abc--
+        extra_headers = {"Content-Type": "multipart/form-data", **(extra_headers or {})}
+        return await self._post(
+            f"/uploads/{upload_id}/parts",
+            body=await async_maybe_transform(body, part_create_params.PartCreateParams),
+            files=files,
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=UploadPart,
+        )
+
+
+class PartsWithRawResponse:
+    def __init__(self, parts: Parts) -> None:
+        self._parts = parts
+
+        self.create = _legacy_response.to_raw_response_wrapper(
+            parts.create,
+        )
+
+
+class AsyncPartsWithRawResponse:
+    def __init__(self, parts: AsyncParts) -> None:
+        self._parts = parts
+
+        self.create = _legacy_response.async_to_raw_response_wrapper(
+            parts.create,
+        )
+
+
+class PartsWithStreamingResponse:
+    def __init__(self, parts: Parts) -> None:
+        self._parts = parts
+
+        self.create = to_streamed_response_wrapper(
+            parts.create,
+        )
+
+
+class AsyncPartsWithStreamingResponse:
+    def __init__(self, parts: AsyncParts) -> None:
+        self._parts = parts
+
+        self.create = async_to_streamed_response_wrapper(
+            parts.create,
+        )

backend/venv/lib/python3.12/site-packages/openai/resources/uploads/uploads.py

@@ -0,0 +1,716 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+import io
+import os
+import logging
+import builtins
+from typing import List, overload
+from pathlib import Path
+
+import anyio
+import httpx
+
+from ... import _legacy_response
+from .parts import (
+    Parts,
+    AsyncParts,
+    PartsWithRawResponse,
+    AsyncPartsWithRawResponse,
+    PartsWithStreamingResponse,
+    AsyncPartsWithStreamingResponse,
+)
+from ...types import FilePurpose, upload_create_params, upload_complete_params
+from ..._types import NOT_GIVEN, Body, Query, Headers, NotGiven
+from ..._utils import (
+    maybe_transform,
+    async_maybe_transform,
+)
+from ..._compat import cached_property
+from ..._resource import SyncAPIResource, AsyncAPIResource
+from ..._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
+from ..._base_client import make_request_options
+from ...types.upload import Upload
+from ...types.file_purpose import FilePurpose
+
+__all__ = ["Uploads", "AsyncUploads"]
+
+
+# 64MB
+DEFAULT_PART_SIZE = 64 * 1024 * 1024
+
+log: logging.Logger = logging.getLogger(__name__)
+
+
+class Uploads(SyncAPIResource):
+    @cached_property
+    def parts(self) -> Parts:
+        return Parts(self._client)
+
+    @cached_property
+    def with_raw_response(self) -> UploadsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return UploadsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> UploadsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return UploadsWithStreamingResponse(self)
+
+    @overload
+    def upload_file_chunked(
+        self,
+        *,
+        file: os.PathLike[str],
+        mime_type: str,
+        purpose: FilePurpose,
+        bytes: int | None = None,
+        part_size: int | None = None,
+        md5: str | NotGiven = NOT_GIVEN,
+    ) -> Upload:
+        """Splits a file into multiple 64MB parts and uploads them sequentially."""
+
+    @overload
+    def upload_file_chunked(
+        self,
+        *,
+        file: bytes,
+        filename: str,
+        bytes: int,
+        mime_type: str,
+        purpose: FilePurpose,
+        part_size: int | None = None,
+        md5: str | NotGiven = NOT_GIVEN,
+    ) -> Upload:
+        """Splits an in-memory file into multiple 64MB parts and uploads them sequentially."""
+
+    def upload_file_chunked(
+        self,
+        *,
+        file: os.PathLike[str] | bytes,
+        mime_type: str,
+        purpose: FilePurpose,
+        filename: str | None = None,
+        bytes: int | None = None,
+        part_size: int | None = None,
+        md5: str | NotGiven = NOT_GIVEN,
+    ) -> Upload:
+        """Splits the given file into multiple parts and uploads them sequentially.
+
+        ```py
+        from pathlib import Path
+
+        client.uploads.upload_file(
+            file=Path("my-paper.pdf"),
+            mime_type="pdf",
+            purpose="assistants",
+        )
+        ```
+        """
+        if isinstance(file, builtins.bytes):
+            if filename is None:
+                raise TypeError("The `filename` argument must be given for in-memory files")
+
+            if bytes is None:
+                raise TypeError("The `bytes` argument must be given for in-memory files")
+        else:
+            if not isinstance(file, Path):
+                file = Path(file)
+
+            if not filename:
+                filename = file.name
+
+            if bytes is None:
+                bytes = file.stat().st_size
+
+        upload = self.create(
+            bytes=bytes,
+            filename=filename,
+            mime_type=mime_type,
+            purpose=purpose,
+        )
+
+        part_ids: list[str] = []
+
+        if part_size is None:
+            part_size = DEFAULT_PART_SIZE
+
+        if isinstance(file, builtins.bytes):
+            buf: io.FileIO | io.BytesIO = io.BytesIO(file)
+        else:
+            buf = io.FileIO(file)
+
+        try:
+            while True:
+                data = buf.read(part_size)
+                if not data:
+                    # EOF
+                    break
+
+                part = self.parts.create(upload_id=upload.id, data=data)
+                log.info("Uploaded part %s for upload %s", part.id, upload.id)
+                part_ids.append(part.id)
+        except Exception:
+            buf.close()
+            raise
+
+        return self.complete(upload_id=upload.id, part_ids=part_ids, md5=md5)
+
+    def create(
+        self,
+        *,
+        bytes: int,
+        filename: str,
+        mime_type: str,
+        purpose: FilePurpose,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Upload:
+        """
+        Creates an intermediate
+        [Upload](https://platform.openai.com/docs/api-reference/uploads/object) object
+        that you can add
+        [Parts](https://platform.openai.com/docs/api-reference/uploads/part-object) to.
+        Currently, an Upload can accept at most 8 GB in total and expires after an hour
+        after you create it.
+
+        Once you complete the Upload, we will create a
+        [File](https://platform.openai.com/docs/api-reference/files/object) object that
+        contains all the parts you uploaded. This File is usable in the rest of our
+        platform as a regular File object.
+
+        For certain `purpose`s, the correct `mime_type` must be specified. Please refer
+        to documentation for the supported MIME types for your use case:
+
+        - [Assistants](https://platform.openai.com/docs/assistants/tools/file-search/supported-files)
+
+        For guidance on the proper filename extensions for each purpose, please follow
+        the documentation on
+        [creating a File](https://platform.openai.com/docs/api-reference/files/create).
+
+        Args:
+          bytes: The number of bytes in the file you are uploading.
+
+          filename: The name of the file to upload.
+
+          mime_type: The MIME type of the file.
+
+              This must fall within the supported MIME types for your file purpose. See the
+              supported MIME types for assistants and vision.
+
+          purpose: The intended purpose of the uploaded file.
+
+              See the
+              [documentation on File purposes](https://platform.openai.com/docs/api-reference/files/create#files-create-purpose).
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        return self._post(
+            "/uploads",
+            body=maybe_transform(
+                {
+                    "bytes": bytes,
+                    "filename": filename,
+                    "mime_type": mime_type,
+                    "purpose": purpose,
+                },
+                upload_create_params.UploadCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=Upload,
+        )
+
+    def cancel(
+        self,
+        upload_id: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Upload:
+        """Cancels the Upload.
+
+        No Parts may be added after an Upload is cancelled.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not upload_id:
+            raise ValueError(f"Expected a non-empty value for `upload_id` but received {upload_id!r}")
+        return self._post(
+            f"/uploads/{upload_id}/cancel",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=Upload,
+        )
+
+    def complete(
+        self,
+        upload_id: str,
+        *,
+        part_ids: List[str],
+        md5: str | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Upload:
+        """
+        Completes the
+        [Upload](https://platform.openai.com/docs/api-reference/uploads/object).
+
+        Within the returned Upload object, there is a nested
+        [File](https://platform.openai.com/docs/api-reference/files/object) object that
+        is ready to use in the rest of the platform.
+
+        You can specify the order of the Parts by passing in an ordered list of the Part
+        IDs.
+
+        The number of bytes uploaded upon completion must match the number of bytes
+        initially specified when creating the Upload object. No Parts may be added after
+        an Upload is completed.
+
+        Args:
+          part_ids: The ordered list of Part IDs.
+
+          md5: The optional md5 checksum for the file contents to verify if the bytes uploaded
+              matches what you expect.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not upload_id:
+            raise ValueError(f"Expected a non-empty value for `upload_id` but received {upload_id!r}")
+        return self._post(
+            f"/uploads/{upload_id}/complete",
+            body=maybe_transform(
+                {
+                    "part_ids": part_ids,
+                    "md5": md5,
+                },
+                upload_complete_params.UploadCompleteParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=Upload,
+        )
+
+
+class AsyncUploads(AsyncAPIResource):
+    @cached_property
+    def parts(self) -> AsyncParts:
+        return AsyncParts(self._client)
+
+    @cached_property
+    def with_raw_response(self) -> AsyncUploadsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return the
+        the raw response object instead of the parsed content.
+
+        For more information, see https://www.github.com/openai/openai-python#accessing-raw-response-data-eg-headers
+        """
+        return AsyncUploadsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncUploadsWithStreamingResponse:
+        """
+        An alternative to `.with_raw_response` that doesn't eagerly read the response body.
+
+        For more information, see https://www.github.com/openai/openai-python#with_streaming_response
+        """
+        return AsyncUploadsWithStreamingResponse(self)
+
+    @overload
+    async def upload_file_chunked(
+        self,
+        *,
+        file: os.PathLike[str],
+        mime_type: str,
+        purpose: FilePurpose,
+        bytes: int | None = None,
+        part_size: int | None = None,
+        md5: str | NotGiven = NOT_GIVEN,
+    ) -> Upload:
+        """Splits a file into multiple 64MB parts and uploads them sequentially."""
+
+    @overload
+    async def upload_file_chunked(
+        self,
+        *,
+        file: bytes,
+        filename: str,
+        bytes: int,
+        mime_type: str,
+        purpose: FilePurpose,
+        part_size: int | None = None,
+        md5: str | NotGiven = NOT_GIVEN,
+    ) -> Upload:
+        """Splits an in-memory file into multiple 64MB parts and uploads them sequentially."""
+
+    async def upload_file_chunked(
+        self,
+        *,
+        file: os.PathLike[str] | bytes,
+        mime_type: str,
+        purpose: FilePurpose,
+        filename: str | None = None,
+        bytes: int | None = None,
+        part_size: int | None = None,
+        md5: str | NotGiven = NOT_GIVEN,
+    ) -> Upload:
+        """Splits the given file into multiple parts and uploads them sequentially.
+
+        ```py
+        from pathlib import Path
+
+        client.uploads.upload_file(
+            file=Path("my-paper.pdf"),
+            mime_type="pdf",
+            purpose="assistants",
+        )
+        ```
+        """
+        if isinstance(file, builtins.bytes):
+            if filename is None:
+                raise TypeError("The `filename` argument must be given for in-memory files")
+
+            if bytes is None:
+                raise TypeError("The `bytes` argument must be given for in-memory files")
+        else:
+            if not isinstance(file, anyio.Path):
+                file = anyio.Path(file)
+
+            if not filename:
+                filename = file.name
+
+            if bytes is None:
+                stat = await file.stat()
+                bytes = stat.st_size
+
+        upload = await self.create(
+            bytes=bytes,
+            filename=filename,
+            mime_type=mime_type,
+            purpose=purpose,
+        )
+
+        part_ids: list[str] = []
+
+        if part_size is None:
+            part_size = DEFAULT_PART_SIZE
+
+        if isinstance(file, anyio.Path):
+            fd = await file.open("rb")
+            async with fd:
+                while True:
+                    data = await fd.read(part_size)
+                    if not data:
+                        # EOF
+                        break
+
+                    part = await self.parts.create(upload_id=upload.id, data=data)
+                    log.info("Uploaded part %s for upload %s", part.id, upload.id)
+                    part_ids.append(part.id)
+        else:
+            buf = io.BytesIO(file)
+
+            try:
+                while True:
+                    data = buf.read(part_size)
+                    if not data:
+                        # EOF
+                        break
+
+                    part = await self.parts.create(upload_id=upload.id, data=data)
+                    log.info("Uploaded part %s for upload %s", part.id, upload.id)
+                    part_ids.append(part.id)
+            except Exception:
+                buf.close()
+                raise
+
+        return await self.complete(upload_id=upload.id, part_ids=part_ids, md5=md5)
+
+    async def create(
+        self,
+        *,
+        bytes: int,
+        filename: str,
+        mime_type: str,
+        purpose: FilePurpose,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Upload:
+        """
+        Creates an intermediate
+        [Upload](https://platform.openai.com/docs/api-reference/uploads/object) object
+        that you can add
+        [Parts](https://platform.openai.com/docs/api-reference/uploads/part-object) to.
+        Currently, an Upload can accept at most 8 GB in total and expires after an hour
+        after you create it.
+
+        Once you complete the Upload, we will create a
+        [File](https://platform.openai.com/docs/api-reference/files/object) object that
+        contains all the parts you uploaded. This File is usable in the rest of our
+        platform as a regular File object.
+
+        For certain `purpose`s, the correct `mime_type` must be specified. Please refer
+        to documentation for the supported MIME types for your use case:
+
+        - [Assistants](https://platform.openai.com/docs/assistants/tools/file-search/supported-files)
+
+        For guidance on the proper filename extensions for each purpose, please follow
+        the documentation on
+        [creating a File](https://platform.openai.com/docs/api-reference/files/create).
+
+        Args:
+          bytes: The number of bytes in the file you are uploading.
+
+          filename: The name of the file to upload.
+
+          mime_type: The MIME type of the file.
+
+              This must fall within the supported MIME types for your file purpose. See the
+              supported MIME types for assistants and vision.
+
+          purpose: The intended purpose of the uploaded file.
+
+              See the
+              [documentation on File purposes](https://platform.openai.com/docs/api-reference/files/create#files-create-purpose).
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        return await self._post(
+            "/uploads",
+            body=await async_maybe_transform(
+                {
+                    "bytes": bytes,
+                    "filename": filename,
+                    "mime_type": mime_type,
+                    "purpose": purpose,
+                },
+                upload_create_params.UploadCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=Upload,
+        )
+
+    async def cancel(
+        self,
+        upload_id: str,
+        *,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Upload:
+        """Cancels the Upload.
+
+        No Parts may be added after an Upload is cancelled.
+
+        Args:
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not upload_id:
+            raise ValueError(f"Expected a non-empty value for `upload_id` but received {upload_id!r}")
+        return await self._post(
+            f"/uploads/{upload_id}/cancel",
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=Upload,
+        )
+
+    async def complete(
+        self,
+        upload_id: str,
+        *,
+        part_ids: List[str],
+        md5: str | NotGiven = NOT_GIVEN,
+        # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
+        # The extra values given here take precedence over values defined on the client or passed to this method.
+        extra_headers: Headers | None = None,
+        extra_query: Query | None = None,
+        extra_body: Body | None = None,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
+    ) -> Upload:
+        """
+        Completes the
+        [Upload](https://platform.openai.com/docs/api-reference/uploads/object).
+
+        Within the returned Upload object, there is a nested
+        [File](https://platform.openai.com/docs/api-reference/files/object) object that
+        is ready to use in the rest of the platform.
+
+        You can specify the order of the Parts by passing in an ordered list of the Part
+        IDs.
+
+        The number of bytes uploaded upon completion must match the number of bytes
+        initially specified when creating the Upload object. No Parts may be added after
+        an Upload is completed.
+
+        Args:
+          part_ids: The ordered list of Part IDs.
+
+          md5: The optional md5 checksum for the file contents to verify if the bytes uploaded
+              matches what you expect.
+
+          extra_headers: Send extra headers
+
+          extra_query: Add additional query parameters to the request
+
+          extra_body: Add additional JSON properties to the request
+
+          timeout: Override the client-level default timeout for this request, in seconds
+        """
+        if not upload_id:
+            raise ValueError(f"Expected a non-empty value for `upload_id` but received {upload_id!r}")
+        return await self._post(
+            f"/uploads/{upload_id}/complete",
+            body=await async_maybe_transform(
+                {
+                    "part_ids": part_ids,
+                    "md5": md5,
+                },
+                upload_complete_params.UploadCompleteParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=Upload,
+        )
+
+
+class UploadsWithRawResponse:
+    def __init__(self, uploads: Uploads) -> None:
+        self._uploads = uploads
+
+        self.create = _legacy_response.to_raw_response_wrapper(
+            uploads.create,
+        )
+        self.cancel = _legacy_response.to_raw_response_wrapper(
+            uploads.cancel,
+        )
+        self.complete = _legacy_response.to_raw_response_wrapper(
+            uploads.complete,
+        )
+
+    @cached_property
+    def parts(self) -> PartsWithRawResponse:
+        return PartsWithRawResponse(self._uploads.parts)
+
+
+class AsyncUploadsWithRawResponse:
+    def __init__(self, uploads: AsyncUploads) -> None:
+        self._uploads = uploads
+
+        self.create = _legacy_response.async_to_raw_response_wrapper(
+            uploads.create,
+        )
+        self.cancel = _legacy_response.async_to_raw_response_wrapper(
+            uploads.cancel,
+        )
+        self.complete = _legacy_response.async_to_raw_response_wrapper(
+            uploads.complete,
+        )
+
+    @cached_property
+    def parts(self) -> AsyncPartsWithRawResponse:
+        return AsyncPartsWithRawResponse(self._uploads.parts)
+
+
+class UploadsWithStreamingResponse:
+    def __init__(self, uploads: Uploads) -> None:
+        self._uploads = uploads
+
+        self.create = to_streamed_response_wrapper(
+            uploads.create,
+        )
+        self.cancel = to_streamed_response_wrapper(
+            uploads.cancel,
+        )
+        self.complete = to_streamed_response_wrapper(
+            uploads.complete,
+        )
+
+    @cached_property
+    def parts(self) -> PartsWithStreamingResponse:
+        return PartsWithStreamingResponse(self._uploads.parts)
+
+
+class AsyncUploadsWithStreamingResponse:
+    def __init__(self, uploads: AsyncUploads) -> None:
+        self._uploads = uploads
+
+        self.create = async_to_streamed_response_wrapper(
+            uploads.create,
+        )
+        self.cancel = async_to_streamed_response_wrapper(
+            uploads.cancel,
+        )
+        self.complete = async_to_streamed_response_wrapper(
+            uploads.complete,
+        )
+
+    @cached_property
+    def parts(self) -> AsyncPartsWithStreamingResponse:
+        return AsyncPartsWithStreamingResponse(self._uploads.parts)