chore: automatic commit 2025-04-30 12:48

venv/lib/python3.11/site-packages/openai/resources/beta/__init__.py

@@ -0,0 +1,47 @@
+# 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,
+)
+
+__all__ = [
+    "Assistants",
+    "AsyncAssistants",
+    "AssistantsWithRawResponse",
+    "AsyncAssistantsWithRawResponse",
+    "AssistantsWithStreamingResponse",
+    "AsyncAssistantsWithStreamingResponse",
+    "Threads",
+    "AsyncThreads",
+    "ThreadsWithRawResponse",
+    "AsyncThreadsWithRawResponse",
+    "ThreadsWithStreamingResponse",
+    "AsyncThreadsWithStreamingResponse",
+    "Beta",
+    "AsyncBeta",
+    "BetaWithRawResponse",
+    "AsyncBetaWithRawResponse",
+    "BetaWithStreamingResponse",
+    "AsyncBetaWithStreamingResponse",
+]

venv/lib/python3.11/site-packages/openai/resources/beta/beta.py

@@ -0,0 +1,175 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+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 .threads.threads import (
+    Threads,
+    AsyncThreads,
+    ThreadsWithRawResponse,
+    AsyncThreadsWithRawResponse,
+    ThreadsWithStreamingResponse,
+    AsyncThreadsWithStreamingResponse,
+)
+from .realtime.realtime import (
+    Realtime,
+    AsyncRealtime,
+    RealtimeWithRawResponse,
+    AsyncRealtimeWithRawResponse,
+    RealtimeWithStreamingResponse,
+    AsyncRealtimeWithStreamingResponse,
+)
+
+__all__ = ["Beta", "AsyncBeta"]
+
+
+class Beta(SyncAPIResource):
+    @cached_property
+    def chat(self) -> Chat:
+        return Chat(self._client)
+
+    @cached_property
+    def realtime(self) -> Realtime:
+        return Realtime(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 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 realtime(self) -> AsyncRealtime:
+        return AsyncRealtime(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 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 realtime(self) -> RealtimeWithRawResponse:
+        return RealtimeWithRawResponse(self._beta.realtime)
+
+    @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 realtime(self) -> AsyncRealtimeWithRawResponse:
+        return AsyncRealtimeWithRawResponse(self._beta.realtime)
+
+    @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 realtime(self) -> RealtimeWithStreamingResponse:
+        return RealtimeWithStreamingResponse(self._beta.realtime)
+
+    @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 realtime(self) -> AsyncRealtimeWithStreamingResponse:
+        return AsyncRealtimeWithStreamingResponse(self._beta.realtime)
+
+    @cached_property
+    def assistants(self) -> AsyncAssistantsWithStreamingResponse:
+        return AsyncAssistantsWithStreamingResponse(self._beta.assistants)
+
+    @cached_property
+    def threads(self) -> AsyncThreadsWithStreamingResponse:
+        return AsyncThreadsWithStreamingResponse(self._beta.threads)

venv/lib/python3.11/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",
+]

venv/lib/python3.11/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)

venv/lib/python3.11/site-packages/openai/resources/beta/chat/completions.py

@@ -0,0 +1,634 @@
+# 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.shared_params import Metadata, ReasoningEffort
+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_audio_param import ChatCompletionAudioParam
+from ....types.chat.chat_completion_message_param import ChatCompletionMessageParam
+from ....types.chat.chat_completion_stream_options_param import ChatCompletionStreamOptionsParam
+from ....types.chat.chat_completion_prediction_content_param import ChatCompletionPredictionContentParam
+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],
+        audio: Optional[ChatCompletionAudioParam] | NotGiven = NOT_GIVEN,
+        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[Metadata] | NotGiven = NOT_GIVEN,
+        modalities: Optional[List[Literal["text", "audio"]]] | NotGiven = NOT_GIVEN,
+        n: Optional[int] | NotGiven = NOT_GIVEN,
+        parallel_tool_calls: bool | NotGiven = NOT_GIVEN,
+        prediction: Optional[ChatCompletionPredictionContentParam] | NotGiven = NOT_GIVEN,
+        presence_penalty: Optional[float] | NotGiven = NOT_GIVEN,
+        reasoning_effort: Optional[ReasoningEffort] | NotGiven = NOT_GIVEN,
+        seed: Optional[int] | NotGiven = NOT_GIVEN,
+        service_tier: Optional[Literal["auto", "default", "flex"]] | NotGiven = NOT_GIVEN,
+        stop: Union[Optional[str], List[str], None] | 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,
+        web_search_options: completion_create_params.WebSearchOptions | 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,
+                    "audio": audio,
+                    "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,
+                    "modalities": modalities,
+                    "n": n,
+                    "parallel_tool_calls": parallel_tool_calls,
+                    "prediction": prediction,
+                    "presence_penalty": presence_penalty,
+                    "reasoning_effort": reasoning_effort,
+                    "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,
+                    "web_search_options": web_search_options,
+                },
+                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],
+        audio: Optional[ChatCompletionAudioParam] | NotGiven = NOT_GIVEN,
+        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[Metadata] | NotGiven = NOT_GIVEN,
+        modalities: Optional[List[Literal["text", "audio"]]] | NotGiven = NOT_GIVEN,
+        n: Optional[int] | NotGiven = NOT_GIVEN,
+        parallel_tool_calls: bool | NotGiven = NOT_GIVEN,
+        prediction: Optional[ChatCompletionPredictionContentParam] | NotGiven = NOT_GIVEN,
+        presence_penalty: Optional[float] | NotGiven = NOT_GIVEN,
+        reasoning_effort: Optional[ReasoningEffort] | NotGiven = NOT_GIVEN,
+        seed: Optional[int] | NotGiven = NOT_GIVEN,
+        service_tier: Optional[Literal["auto", "default", "flex"]] | NotGiven = NOT_GIVEN,
+        stop: Union[Optional[str], List[str], None] | 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,
+        web_search_options: completion_create_params.WebSearchOptions | 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,
+            audio=audio,
+            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,
+            modalities=modalities,
+            n=n,
+            parallel_tool_calls=parallel_tool_calls,
+            prediction=prediction,
+            presence_penalty=presence_penalty,
+            reasoning_effort=reasoning_effort,
+            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,
+            web_search_options=web_search_options,
+            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],
+        audio: Optional[ChatCompletionAudioParam] | NotGiven = NOT_GIVEN,
+        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[Metadata] | NotGiven = NOT_GIVEN,
+        modalities: Optional[List[Literal["text", "audio"]]] | NotGiven = NOT_GIVEN,
+        n: Optional[int] | NotGiven = NOT_GIVEN,
+        parallel_tool_calls: bool | NotGiven = NOT_GIVEN,
+        prediction: Optional[ChatCompletionPredictionContentParam] | NotGiven = NOT_GIVEN,
+        presence_penalty: Optional[float] | NotGiven = NOT_GIVEN,
+        reasoning_effort: Optional[ReasoningEffort] | NotGiven = NOT_GIVEN,
+        seed: Optional[int] | NotGiven = NOT_GIVEN,
+        service_tier: Optional[Literal["auto", "default", "flex"]] | NotGiven = NOT_GIVEN,
+        stop: Union[Optional[str], List[str], None] | 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,
+        web_search_options: completion_create_params.WebSearchOptions | 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,
+                    "audio": audio,
+                    "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,
+                    "modalities": modalities,
+                    "n": n,
+                    "parallel_tool_calls": parallel_tool_calls,
+                    "prediction": prediction,
+                    "presence_penalty": presence_penalty,
+                    "reasoning_effort": reasoning_effort,
+                    "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,
+                    "web_search_options": web_search_options,
+                },
+                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],
+        audio: Optional[ChatCompletionAudioParam] | NotGiven = NOT_GIVEN,
+        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[Metadata] | NotGiven = NOT_GIVEN,
+        modalities: Optional[List[Literal["text", "audio"]]] | NotGiven = NOT_GIVEN,
+        n: Optional[int] | NotGiven = NOT_GIVEN,
+        parallel_tool_calls: bool | NotGiven = NOT_GIVEN,
+        prediction: Optional[ChatCompletionPredictionContentParam] | NotGiven = NOT_GIVEN,
+        presence_penalty: Optional[float] | NotGiven = NOT_GIVEN,
+        reasoning_effort: Optional[ReasoningEffort] | NotGiven = NOT_GIVEN,
+        seed: Optional[int] | NotGiven = NOT_GIVEN,
+        service_tier: Optional[Literal["auto", "default", "flex"]] | NotGiven = NOT_GIVEN,
+        stop: Union[Optional[str], List[str], None] | 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,
+        web_search_options: completion_create_params.WebSearchOptions | 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,
+            audio=audio,
+            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,
+            modalities=modalities,
+            n=n,
+            parallel_tool_calls=parallel_tool_calls,
+            prediction=prediction,
+            presence_penalty=presence_penalty,
+            reasoning_effort=reasoning_effort,
+            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,
+            web_search_options=web_search_options,
+        )
+        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,
+        )

venv/lib/python3.11/site-packages/openai/resources/beta/realtime/__init__.py

@@ -0,0 +1,47 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from .realtime import (
+    Realtime,
+    AsyncRealtime,
+    RealtimeWithRawResponse,
+    AsyncRealtimeWithRawResponse,
+    RealtimeWithStreamingResponse,
+    AsyncRealtimeWithStreamingResponse,
+)
+from .sessions import (
+    Sessions,
+    AsyncSessions,
+    SessionsWithRawResponse,
+    AsyncSessionsWithRawResponse,
+    SessionsWithStreamingResponse,
+    AsyncSessionsWithStreamingResponse,
+)
+from .transcription_sessions import (
+    TranscriptionSessions,
+    AsyncTranscriptionSessions,
+    TranscriptionSessionsWithRawResponse,
+    AsyncTranscriptionSessionsWithRawResponse,
+    TranscriptionSessionsWithStreamingResponse,
+    AsyncTranscriptionSessionsWithStreamingResponse,
+)
+
+__all__ = [
+    "Sessions",
+    "AsyncSessions",
+    "SessionsWithRawResponse",
+    "AsyncSessionsWithRawResponse",
+    "SessionsWithStreamingResponse",
+    "AsyncSessionsWithStreamingResponse",
+    "TranscriptionSessions",
+    "AsyncTranscriptionSessions",
+    "TranscriptionSessionsWithRawResponse",
+    "AsyncTranscriptionSessionsWithRawResponse",
+    "TranscriptionSessionsWithStreamingResponse",
+    "AsyncTranscriptionSessionsWithStreamingResponse",
+    "Realtime",
+    "AsyncRealtime",
+    "RealtimeWithRawResponse",
+    "AsyncRealtimeWithRawResponse",
+    "RealtimeWithStreamingResponse",
+    "AsyncRealtimeWithStreamingResponse",
+]

venv/lib/python3.11/site-packages/openai/resources/beta/realtime/sessions.py

@@ -0,0 +1,388 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import List, Union, Iterable
+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 ...._base_client import make_request_options
+from ....types.beta.realtime import session_create_params
+from ....types.beta.realtime.session_create_response import SessionCreateResponse
+
+__all__ = ["Sessions", "AsyncSessions"]
+
+
+class Sessions(SyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> SessionsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return
+        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 SessionsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> SessionsWithStreamingResponse:
+        """
+        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 SessionsWithStreamingResponse(self)
+
+    def create(
+        self,
+        *,
+        input_audio_format: Literal["pcm16", "g711_ulaw", "g711_alaw"] | NotGiven = NOT_GIVEN,
+        input_audio_noise_reduction: session_create_params.InputAudioNoiseReduction | NotGiven = NOT_GIVEN,
+        input_audio_transcription: session_create_params.InputAudioTranscription | NotGiven = NOT_GIVEN,
+        instructions: str | NotGiven = NOT_GIVEN,
+        max_response_output_tokens: Union[int, Literal["inf"]] | NotGiven = NOT_GIVEN,
+        modalities: List[Literal["text", "audio"]] | NotGiven = NOT_GIVEN,
+        model: Literal[
+            "gpt-4o-realtime-preview",
+            "gpt-4o-realtime-preview-2024-10-01",
+            "gpt-4o-realtime-preview-2024-12-17",
+            "gpt-4o-mini-realtime-preview",
+            "gpt-4o-mini-realtime-preview-2024-12-17",
+        ]
+        | NotGiven = NOT_GIVEN,
+        output_audio_format: Literal["pcm16", "g711_ulaw", "g711_alaw"] | NotGiven = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        tool_choice: str | NotGiven = NOT_GIVEN,
+        tools: Iterable[session_create_params.Tool] | NotGiven = NOT_GIVEN,
+        turn_detection: session_create_params.TurnDetection | NotGiven = NOT_GIVEN,
+        voice: Union[
+            str, Literal["alloy", "ash", "ballad", "coral", "echo", "fable", "onyx", "nova", "sage", "shimmer", "verse"]
+        ]
+        | 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,
+    ) -> SessionCreateResponse:
+        """
+        Create an ephemeral API token for use in client-side applications with the
+        Realtime API. Can be configured with the same session parameters as the
+        `session.update` client event.
+
+        It responds with a session object, plus a `client_secret` key which contains a
+        usable ephemeral API token that can be used to authenticate browser clients for
+        the Realtime API.
+
+        Args:
+          input_audio_format: The format of input audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`. For
+              `pcm16`, input audio must be 16-bit PCM at a 24kHz sample rate, single channel
+              (mono), and little-endian byte order.
+
+          input_audio_noise_reduction: Configuration for input audio noise reduction. This can be set to `null` to turn
+              off. Noise reduction filters audio added to the input audio buffer before it is
+              sent to VAD and the model. Filtering the audio can improve VAD and turn
+              detection accuracy (reducing false positives) and model performance by improving
+              perception of the input audio.
+
+          input_audio_transcription: Configuration for input audio transcription, defaults to off and can be set to
+              `null` to turn off once on. Input audio transcription is not native to the
+              model, since the model consumes audio directly. Transcription runs
+              asynchronously through
+              [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription)
+              and should be treated as guidance of input audio content rather than precisely
+              what the model heard. The client can optionally set the language and prompt for
+              transcription, these offer additional guidance to the transcription service.
+
+          instructions: The default system instructions (i.e. system message) prepended to model calls.
+              This field allows the client to guide the model on desired responses. The model
+              can be instructed on response content and format, (e.g. "be extremely succinct",
+              "act friendly", "here are examples of good responses") and on audio behavior
+              (e.g. "talk quickly", "inject emotion into your voice", "laugh frequently"). The
+              instructions are not guaranteed to be followed by the model, but they provide
+              guidance to the model on the desired behavior.
+
+              Note that the server sets default instructions which will be used if this field
+              is not set and are visible in the `session.created` event at the start of the
+              session.
+
+          max_response_output_tokens: Maximum number of output tokens for a single assistant response, inclusive of
+              tool calls. Provide an integer between 1 and 4096 to limit output tokens, or
+              `inf` for the maximum available tokens for a given model. Defaults to `inf`.
+
+          modalities: The set of modalities the model can respond with. To disable audio, set this to
+              ["text"].
+
+          model: The Realtime model used for this session.
+
+          output_audio_format: The format of output audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`.
+              For `pcm16`, output audio is sampled at a rate of 24kHz.
+
+          temperature: Sampling temperature for the model, limited to [0.6, 1.2]. For audio models a
+              temperature of 0.8 is highly recommended for best performance.
+
+          tool_choice: How the model chooses tools. Options are `auto`, `none`, `required`, or specify
+              a function.
+
+          tools: Tools (functions) available to the model.
+
+          turn_detection: Configuration for turn detection, ether Server VAD or Semantic VAD. This can be
+              set to `null` to turn off, in which case the client must manually trigger model
+              response. Server VAD means that the model will detect the start and end of
+              speech based on audio volume and respond at the end of user speech. Semantic VAD
+              is more advanced and uses a turn detection model (in conjuction with VAD) to
+              semantically estimate whether the user has finished speaking, then dynamically
+              sets a timeout based on this probability. For example, if user audio trails off
+              with "uhhm", the model will score a low probability of turn end and wait longer
+              for the user to continue speaking. This can be useful for more natural
+              conversations, but may have a higher latency.
+
+          voice: The voice the model uses to respond. Voice cannot be changed during the session
+              once the model has responded with audio at least once. Current voice options are
+              `alloy`, `ash`, `ballad`, `coral`, `echo`, `fable`, `onyx`, `nova`, `sage`,
+              `shimmer`, and `verse`.
+
+          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(
+            "/realtime/sessions",
+            body=maybe_transform(
+                {
+                    "input_audio_format": input_audio_format,
+                    "input_audio_noise_reduction": input_audio_noise_reduction,
+                    "input_audio_transcription": input_audio_transcription,
+                    "instructions": instructions,
+                    "max_response_output_tokens": max_response_output_tokens,
+                    "modalities": modalities,
+                    "model": model,
+                    "output_audio_format": output_audio_format,
+                    "temperature": temperature,
+                    "tool_choice": tool_choice,
+                    "tools": tools,
+                    "turn_detection": turn_detection,
+                    "voice": voice,
+                },
+                session_create_params.SessionCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=SessionCreateResponse,
+        )
+
+
+class AsyncSessions(AsyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> AsyncSessionsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return
+        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 AsyncSessionsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncSessionsWithStreamingResponse:
+        """
+        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 AsyncSessionsWithStreamingResponse(self)
+
+    async def create(
+        self,
+        *,
+        input_audio_format: Literal["pcm16", "g711_ulaw", "g711_alaw"] | NotGiven = NOT_GIVEN,
+        input_audio_noise_reduction: session_create_params.InputAudioNoiseReduction | NotGiven = NOT_GIVEN,
+        input_audio_transcription: session_create_params.InputAudioTranscription | NotGiven = NOT_GIVEN,
+        instructions: str | NotGiven = NOT_GIVEN,
+        max_response_output_tokens: Union[int, Literal["inf"]] | NotGiven = NOT_GIVEN,
+        modalities: List[Literal["text", "audio"]] | NotGiven = NOT_GIVEN,
+        model: Literal[
+            "gpt-4o-realtime-preview",
+            "gpt-4o-realtime-preview-2024-10-01",
+            "gpt-4o-realtime-preview-2024-12-17",
+            "gpt-4o-mini-realtime-preview",
+            "gpt-4o-mini-realtime-preview-2024-12-17",
+        ]
+        | NotGiven = NOT_GIVEN,
+        output_audio_format: Literal["pcm16", "g711_ulaw", "g711_alaw"] | NotGiven = NOT_GIVEN,
+        temperature: float | NotGiven = NOT_GIVEN,
+        tool_choice: str | NotGiven = NOT_GIVEN,
+        tools: Iterable[session_create_params.Tool] | NotGiven = NOT_GIVEN,
+        turn_detection: session_create_params.TurnDetection | NotGiven = NOT_GIVEN,
+        voice: Union[
+            str, Literal["alloy", "ash", "ballad", "coral", "echo", "fable", "onyx", "nova", "sage", "shimmer", "verse"]
+        ]
+        | 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,
+    ) -> SessionCreateResponse:
+        """
+        Create an ephemeral API token for use in client-side applications with the
+        Realtime API. Can be configured with the same session parameters as the
+        `session.update` client event.
+
+        It responds with a session object, plus a `client_secret` key which contains a
+        usable ephemeral API token that can be used to authenticate browser clients for
+        the Realtime API.
+
+        Args:
+          input_audio_format: The format of input audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`. For
+              `pcm16`, input audio must be 16-bit PCM at a 24kHz sample rate, single channel
+              (mono), and little-endian byte order.
+
+          input_audio_noise_reduction: Configuration for input audio noise reduction. This can be set to `null` to turn
+              off. Noise reduction filters audio added to the input audio buffer before it is
+              sent to VAD and the model. Filtering the audio can improve VAD and turn
+              detection accuracy (reducing false positives) and model performance by improving
+              perception of the input audio.
+
+          input_audio_transcription: Configuration for input audio transcription, defaults to off and can be set to
+              `null` to turn off once on. Input audio transcription is not native to the
+              model, since the model consumes audio directly. Transcription runs
+              asynchronously through
+              [the /audio/transcriptions endpoint](https://platform.openai.com/docs/api-reference/audio/createTranscription)
+              and should be treated as guidance of input audio content rather than precisely
+              what the model heard. The client can optionally set the language and prompt for
+              transcription, these offer additional guidance to the transcription service.
+
+          instructions: The default system instructions (i.e. system message) prepended to model calls.
+              This field allows the client to guide the model on desired responses. The model
+              can be instructed on response content and format, (e.g. "be extremely succinct",
+              "act friendly", "here are examples of good responses") and on audio behavior
+              (e.g. "talk quickly", "inject emotion into your voice", "laugh frequently"). The
+              instructions are not guaranteed to be followed by the model, but they provide
+              guidance to the model on the desired behavior.
+
+              Note that the server sets default instructions which will be used if this field
+              is not set and are visible in the `session.created` event at the start of the
+              session.
+
+          max_response_output_tokens: Maximum number of output tokens for a single assistant response, inclusive of
+              tool calls. Provide an integer between 1 and 4096 to limit output tokens, or
+              `inf` for the maximum available tokens for a given model. Defaults to `inf`.
+
+          modalities: The set of modalities the model can respond with. To disable audio, set this to
+              ["text"].
+
+          model: The Realtime model used for this session.
+
+          output_audio_format: The format of output audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`.
+              For `pcm16`, output audio is sampled at a rate of 24kHz.
+
+          temperature: Sampling temperature for the model, limited to [0.6, 1.2]. For audio models a
+              temperature of 0.8 is highly recommended for best performance.
+
+          tool_choice: How the model chooses tools. Options are `auto`, `none`, `required`, or specify
+              a function.
+
+          tools: Tools (functions) available to the model.
+
+          turn_detection: Configuration for turn detection, ether Server VAD or Semantic VAD. This can be
+              set to `null` to turn off, in which case the client must manually trigger model
+              response. Server VAD means that the model will detect the start and end of
+              speech based on audio volume and respond at the end of user speech. Semantic VAD
+              is more advanced and uses a turn detection model (in conjuction with VAD) to
+              semantically estimate whether the user has finished speaking, then dynamically
+              sets a timeout based on this probability. For example, if user audio trails off
+              with "uhhm", the model will score a low probability of turn end and wait longer
+              for the user to continue speaking. This can be useful for more natural
+              conversations, but may have a higher latency.
+
+          voice: The voice the model uses to respond. Voice cannot be changed during the session
+              once the model has responded with audio at least once. Current voice options are
+              `alloy`, `ash`, `ballad`, `coral`, `echo`, `fable`, `onyx`, `nova`, `sage`,
+              `shimmer`, and `verse`.
+
+          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(
+            "/realtime/sessions",
+            body=await async_maybe_transform(
+                {
+                    "input_audio_format": input_audio_format,
+                    "input_audio_noise_reduction": input_audio_noise_reduction,
+                    "input_audio_transcription": input_audio_transcription,
+                    "instructions": instructions,
+                    "max_response_output_tokens": max_response_output_tokens,
+                    "modalities": modalities,
+                    "model": model,
+                    "output_audio_format": output_audio_format,
+                    "temperature": temperature,
+                    "tool_choice": tool_choice,
+                    "tools": tools,
+                    "turn_detection": turn_detection,
+                    "voice": voice,
+                },
+                session_create_params.SessionCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=SessionCreateResponse,
+        )
+
+
+class SessionsWithRawResponse:
+    def __init__(self, sessions: Sessions) -> None:
+        self._sessions = sessions
+
+        self.create = _legacy_response.to_raw_response_wrapper(
+            sessions.create,
+        )
+
+
+class AsyncSessionsWithRawResponse:
+    def __init__(self, sessions: AsyncSessions) -> None:
+        self._sessions = sessions
+
+        self.create = _legacy_response.async_to_raw_response_wrapper(
+            sessions.create,
+        )
+
+
+class SessionsWithStreamingResponse:
+    def __init__(self, sessions: Sessions) -> None:
+        self._sessions = sessions
+
+        self.create = to_streamed_response_wrapper(
+            sessions.create,
+        )
+
+
+class AsyncSessionsWithStreamingResponse:
+    def __init__(self, sessions: AsyncSessions) -> None:
+        self._sessions = sessions
+
+        self.create = async_to_streamed_response_wrapper(
+            sessions.create,
+        )

venv/lib/python3.11/site-packages/openai/resources/beta/realtime/transcription_sessions.py

@@ -0,0 +1,274 @@
+# 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 ...._base_client import make_request_options
+from ....types.beta.realtime import transcription_session_create_params
+from ....types.beta.realtime.transcription_session import TranscriptionSession
+
+__all__ = ["TranscriptionSessions", "AsyncTranscriptionSessions"]
+
+
+class TranscriptionSessions(SyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> TranscriptionSessionsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return
+        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 TranscriptionSessionsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> TranscriptionSessionsWithStreamingResponse:
+        """
+        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 TranscriptionSessionsWithStreamingResponse(self)
+
+    def create(
+        self,
+        *,
+        include: List[str] | NotGiven = NOT_GIVEN,
+        input_audio_format: Literal["pcm16", "g711_ulaw", "g711_alaw"] | NotGiven = NOT_GIVEN,
+        input_audio_noise_reduction: transcription_session_create_params.InputAudioNoiseReduction
+        | NotGiven = NOT_GIVEN,
+        input_audio_transcription: transcription_session_create_params.InputAudioTranscription | NotGiven = NOT_GIVEN,
+        modalities: List[Literal["text", "audio"]] | NotGiven = NOT_GIVEN,
+        turn_detection: transcription_session_create_params.TurnDetection | 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,
+    ) -> TranscriptionSession:
+        """
+        Create an ephemeral API token for use in client-side applications with the
+        Realtime API specifically for realtime transcriptions. Can be configured with
+        the same session parameters as the `transcription_session.update` client event.
+
+        It responds with a session object, plus a `client_secret` key which contains a
+        usable ephemeral API token that can be used to authenticate browser clients for
+        the Realtime API.
+
+        Args:
+          include:
+              The set of items to include in the transcription. Current available items are:
+
+              - `item.input_audio_transcription.logprobs`
+
+          input_audio_format: The format of input audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`. For
+              `pcm16`, input audio must be 16-bit PCM at a 24kHz sample rate, single channel
+              (mono), and little-endian byte order.
+
+          input_audio_noise_reduction: Configuration for input audio noise reduction. This can be set to `null` to turn
+              off. Noise reduction filters audio added to the input audio buffer before it is
+              sent to VAD and the model. Filtering the audio can improve VAD and turn
+              detection accuracy (reducing false positives) and model performance by improving
+              perception of the input audio.
+
+          input_audio_transcription: Configuration for input audio transcription. The client can optionally set the
+              language and prompt for transcription, these offer additional guidance to the
+              transcription service.
+
+          modalities: The set of modalities the model can respond with. To disable audio, set this to
+              ["text"].
+
+          turn_detection: Configuration for turn detection, ether Server VAD or Semantic VAD. This can be
+              set to `null` to turn off, in which case the client must manually trigger model
+              response. Server VAD means that the model will detect the start and end of
+              speech based on audio volume and respond at the end of user speech. Semantic VAD
+              is more advanced and uses a turn detection model (in conjuction with VAD) to
+              semantically estimate whether the user has finished speaking, then dynamically
+              sets a timeout based on this probability. For example, if user audio trails off
+              with "uhhm", the model will score a low probability of turn end and wait longer
+              for the user to continue speaking. This can be useful for more natural
+              conversations, but may have a higher 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
+        """
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return self._post(
+            "/realtime/transcription_sessions",
+            body=maybe_transform(
+                {
+                    "include": include,
+                    "input_audio_format": input_audio_format,
+                    "input_audio_noise_reduction": input_audio_noise_reduction,
+                    "input_audio_transcription": input_audio_transcription,
+                    "modalities": modalities,
+                    "turn_detection": turn_detection,
+                },
+                transcription_session_create_params.TranscriptionSessionCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=TranscriptionSession,
+        )
+
+
+class AsyncTranscriptionSessions(AsyncAPIResource):
+    @cached_property
+    def with_raw_response(self) -> AsyncTranscriptionSessionsWithRawResponse:
+        """
+        This property can be used as a prefix for any HTTP method call to return
+        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 AsyncTranscriptionSessionsWithRawResponse(self)
+
+    @cached_property
+    def with_streaming_response(self) -> AsyncTranscriptionSessionsWithStreamingResponse:
+        """
+        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 AsyncTranscriptionSessionsWithStreamingResponse(self)
+
+    async def create(
+        self,
+        *,
+        include: List[str] | NotGiven = NOT_GIVEN,
+        input_audio_format: Literal["pcm16", "g711_ulaw", "g711_alaw"] | NotGiven = NOT_GIVEN,
+        input_audio_noise_reduction: transcription_session_create_params.InputAudioNoiseReduction
+        | NotGiven = NOT_GIVEN,
+        input_audio_transcription: transcription_session_create_params.InputAudioTranscription | NotGiven = NOT_GIVEN,
+        modalities: List[Literal["text", "audio"]] | NotGiven = NOT_GIVEN,
+        turn_detection: transcription_session_create_params.TurnDetection | 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,
+    ) -> TranscriptionSession:
+        """
+        Create an ephemeral API token for use in client-side applications with the
+        Realtime API specifically for realtime transcriptions. Can be configured with
+        the same session parameters as the `transcription_session.update` client event.
+
+        It responds with a session object, plus a `client_secret` key which contains a
+        usable ephemeral API token that can be used to authenticate browser clients for
+        the Realtime API.
+
+        Args:
+          include:
+              The set of items to include in the transcription. Current available items are:
+
+              - `item.input_audio_transcription.logprobs`
+
+          input_audio_format: The format of input audio. Options are `pcm16`, `g711_ulaw`, or `g711_alaw`. For
+              `pcm16`, input audio must be 16-bit PCM at a 24kHz sample rate, single channel
+              (mono), and little-endian byte order.
+
+          input_audio_noise_reduction: Configuration for input audio noise reduction. This can be set to `null` to turn
+              off. Noise reduction filters audio added to the input audio buffer before it is
+              sent to VAD and the model. Filtering the audio can improve VAD and turn
+              detection accuracy (reducing false positives) and model performance by improving
+              perception of the input audio.
+
+          input_audio_transcription: Configuration for input audio transcription. The client can optionally set the
+              language and prompt for transcription, these offer additional guidance to the
+              transcription service.
+
+          modalities: The set of modalities the model can respond with. To disable audio, set this to
+              ["text"].
+
+          turn_detection: Configuration for turn detection, ether Server VAD or Semantic VAD. This can be
+              set to `null` to turn off, in which case the client must manually trigger model
+              response. Server VAD means that the model will detect the start and end of
+              speech based on audio volume and respond at the end of user speech. Semantic VAD
+              is more advanced and uses a turn detection model (in conjuction with VAD) to
+              semantically estimate whether the user has finished speaking, then dynamically
+              sets a timeout based on this probability. For example, if user audio trails off
+              with "uhhm", the model will score a low probability of turn end and wait longer
+              for the user to continue speaking. This can be useful for more natural
+              conversations, but may have a higher 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
+        """
+        extra_headers = {"OpenAI-Beta": "assistants=v2", **(extra_headers or {})}
+        return await self._post(
+            "/realtime/transcription_sessions",
+            body=await async_maybe_transform(
+                {
+                    "include": include,
+                    "input_audio_format": input_audio_format,
+                    "input_audio_noise_reduction": input_audio_noise_reduction,
+                    "input_audio_transcription": input_audio_transcription,
+                    "modalities": modalities,
+                    "turn_detection": turn_detection,
+                },
+                transcription_session_create_params.TranscriptionSessionCreateParams,
+            ),
+            options=make_request_options(
+                extra_headers=extra_headers, extra_query=extra_query, extra_body=extra_body, timeout=timeout
+            ),
+            cast_to=TranscriptionSession,
+        )
+
+
+class TranscriptionSessionsWithRawResponse:
+    def __init__(self, transcription_sessions: TranscriptionSessions) -> None:
+        self._transcription_sessions = transcription_sessions
+
+        self.create = _legacy_response.to_raw_response_wrapper(
+            transcription_sessions.create,
+        )
+
+
+class AsyncTranscriptionSessionsWithRawResponse:
+    def __init__(self, transcription_sessions: AsyncTranscriptionSessions) -> None:
+        self._transcription_sessions = transcription_sessions
+
+        self.create = _legacy_response.async_to_raw_response_wrapper(
+            transcription_sessions.create,
+        )
+
+
+class TranscriptionSessionsWithStreamingResponse:
+    def __init__(self, transcription_sessions: TranscriptionSessions) -> None:
+        self._transcription_sessions = transcription_sessions
+
+        self.create = to_streamed_response_wrapper(
+            transcription_sessions.create,
+        )
+
+
+class AsyncTranscriptionSessionsWithStreamingResponse:
+    def __init__(self, transcription_sessions: AsyncTranscriptionSessions) -> None:
+        self._transcription_sessions = transcription_sessions
+
+        self.create = async_to_streamed_response_wrapper(
+            transcription_sessions.create,
+        )

venv/lib/python3.11/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",
+]

venv/lib/python3.11/site-packages/openai/resources/beta/threads/messages.py

@@ -0,0 +1,667 @@
+# 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.shared_params.metadata import Metadata
+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 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[Metadata] | 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, and
+              querying for objects via API or the dashboard.
+
+              Keys are strings with a maximum length of 64 characters. Values are strings with
+              a maximum length of 512 characters.
+
+          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[Metadata] | 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, and
+              querying for objects via API or the dashboard.
+
+              Keys are strings with a maximum length of 64 characters. Values are strings with
+              a maximum length of 512 characters.
+
+          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,
+              starting 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 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[Metadata] | 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, and
+              querying for objects via API or the dashboard.
+
+              Keys are strings with a maximum length of 64 characters. Values are strings with
+              a maximum length of 512 characters.
+
+          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[Metadata] | 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, and
+              querying for objects via API or the dashboard.
+
+              Keys are strings with a maximum length of 64 characters. Values are strings with
+              a maximum length of 512 characters.
+
+          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,
+              starting 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,
+        )

venv/lib/python3.11/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",
+]

venv/lib/python3.11/site-packages/openai/resources/beta/threads/runs/steps.py

@@ -0,0 +1,378 @@
+# 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 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,
+              starting 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 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,
+              starting 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,
+        )