feat(api): add /v1/responses and built-in tools (openai#2177)

[platform.openai.com/docs/changelog](http://platform.openai.com/docs/changelog)

Author: stainless-app[bot]

.stats.yml

@@ -1,2 +1,2 @@
-configured_endpoints: 74
-openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/openai-b524aed1c2c5c928aa4e2c546f5dbb364e7b4d5027daf05e42e210b05a97c3c6.yml
+configured_endpoints: 81
+openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/openai-be834d63e326a82494e819085137f5eb15866f3fc787db1f3afe7168d419e18a.yml

api.md

@@ -37,7 +37,9 @@
 from .resources.chat import chat
 from .resources.audio import audio
 from .resources.uploads import uploads
+from .resources.responses import responses
 from .resources.fine_tuning import fine_tuning
+from .resources.vector_stores import vector_stores
 
 __all__ = ["Timeout", "Transport", "ProxiesTypes", "RequestOptions", "OpenAI", "AsyncOpenAI", "Client", "AsyncClient"]
 
@@ -52,9 +54,11 @@ class OpenAI(SyncAPIClient):
     moderations: moderations.Moderations
     models: models.Models
     fine_tuning: fine_tuning.FineTuning
+    vector_stores: vector_stores.VectorStores
     beta: beta.Beta
     batches: batches.Batches
     uploads: uploads.Uploads
+    responses: responses.Responses
     with_raw_response: OpenAIWithRawResponse
     with_streaming_response: OpenAIWithStreamedResponse
 
@@ -149,9 +153,11 @@ def __init__(
         self.moderations = moderations.Moderations(self)
         self.models = models.Models(self)
         self.fine_tuning = fine_tuning.FineTuning(self)
+        self.vector_stores = vector_stores.VectorStores(self)
         self.beta = beta.Beta(self)
         self.batches = batches.Batches(self)
         self.uploads = uploads.Uploads(self)
+        self.responses = responses.Responses(self)
         self.with_raw_response = OpenAIWithRawResponse(self)
         self.with_streaming_response = OpenAIWithStreamedResponse(self)
 
@@ -279,9 +285,11 @@ class AsyncOpenAI(AsyncAPIClient):
     moderations: moderations.AsyncModerations
     models: models.AsyncModels
     fine_tuning: fine_tuning.AsyncFineTuning
+    vector_stores: vector_stores.AsyncVectorStores
     beta: beta.AsyncBeta
     batches: batches.AsyncBatches
     uploads: uploads.AsyncUploads
+    responses: responses.AsyncResponses
     with_raw_response: AsyncOpenAIWithRawResponse
     with_streaming_response: AsyncOpenAIWithStreamedResponse
 
@@ -376,9 +384,11 @@ def __init__(
         self.moderations = moderations.AsyncModerations(self)
         self.models = models.AsyncModels(self)
         self.fine_tuning = fine_tuning.AsyncFineTuning(self)
+        self.vector_stores = vector_stores.AsyncVectorStores(self)
         self.beta = beta.AsyncBeta(self)
         self.batches = batches.AsyncBatches(self)
         self.uploads = uploads.AsyncUploads(self)
+        self.responses = responses.AsyncResponses(self)
         self.with_raw_response = AsyncOpenAIWithRawResponse(self)
         self.with_streaming_response = AsyncOpenAIWithStreamedResponse(self)
 
@@ -507,9 +517,11 @@ def __init__(self, client: OpenAI) -> None:
         self.moderations = moderations.ModerationsWithRawResponse(client.moderations)
         self.models = models.ModelsWithRawResponse(client.models)
         self.fine_tuning = fine_tuning.FineTuningWithRawResponse(client.fine_tuning)
+        self.vector_stores = vector_stores.VectorStoresWithRawResponse(client.vector_stores)
         self.beta = beta.BetaWithRawResponse(client.beta)
         self.batches = batches.BatchesWithRawResponse(client.batches)
         self.uploads = uploads.UploadsWithRawResponse(client.uploads)
+        self.responses = responses.ResponsesWithRawResponse(client.responses)
 
 
 class AsyncOpenAIWithRawResponse:
@@ -523,9 +535,11 @@ def __init__(self, client: AsyncOpenAI) -> None:
         self.moderations = moderations.AsyncModerationsWithRawResponse(client.moderations)
         self.models = models.AsyncModelsWithRawResponse(client.models)
         self.fine_tuning = fine_tuning.AsyncFineTuningWithRawResponse(client.fine_tuning)
+        self.vector_stores = vector_stores.AsyncVectorStoresWithRawResponse(client.vector_stores)
         self.beta = beta.AsyncBetaWithRawResponse(client.beta)
         self.batches = batches.AsyncBatchesWithRawResponse(client.batches)
         self.uploads = uploads.AsyncUploadsWithRawResponse(client.uploads)
+        self.responses = responses.AsyncResponsesWithRawResponse(client.responses)
 
 
 class OpenAIWithStreamedResponse:
@@ -539,9 +553,11 @@ def __init__(self, client: OpenAI) -> None:
         self.moderations = moderations.ModerationsWithStreamingResponse(client.moderations)
         self.models = models.ModelsWithStreamingResponse(client.models)
         self.fine_tuning = fine_tuning.FineTuningWithStreamingResponse(client.fine_tuning)
+        self.vector_stores = vector_stores.VectorStoresWithStreamingResponse(client.vector_stores)
         self.beta = beta.BetaWithStreamingResponse(client.beta)
         self.batches = batches.BatchesWithStreamingResponse(client.batches)
         self.uploads = uploads.UploadsWithStreamingResponse(client.uploads)
+        self.responses = responses.ResponsesWithStreamingResponse(client.responses)
 
 
 class AsyncOpenAIWithStreamedResponse:
@@ -555,9 +571,11 @@ def __init__(self, client: AsyncOpenAI) -> None:
         self.moderations = moderations.AsyncModerationsWithStreamingResponse(client.moderations)
         self.models = models.AsyncModelsWithStreamingResponse(client.models)
         self.fine_tuning = fine_tuning.AsyncFineTuningWithStreamingResponse(client.fine_tuning)
+        self.vector_stores = vector_stores.AsyncVectorStoresWithStreamingResponse(client.vector_stores)
         self.beta = beta.AsyncBetaWithStreamingResponse(client.beta)
         self.batches = batches.AsyncBatchesWithStreamingResponse(client.batches)
         self.uploads = uploads.AsyncUploadsWithStreamingResponse(client.uploads)
+        self.responses = responses.AsyncResponsesWithStreamingResponse(client.responses)
 
 
 Client = OpenAI

src/openai/_client.py

@@ -64,6 +64,14 @@
     UploadsWithStreamingResponse,
     AsyncUploadsWithStreamingResponse,
 )
+from .responses import (
+    Responses,
+    AsyncResponses,
+    ResponsesWithRawResponse,
+    AsyncResponsesWithRawResponse,
+    ResponsesWithStreamingResponse,
+    AsyncResponsesWithStreamingResponse,
+)
 from .embeddings import (
     Embeddings,
     AsyncEmbeddings,
@@ -96,6 +104,14 @@
     ModerationsWithStreamingResponse,
     AsyncModerationsWithStreamingResponse,
 )
+from .vector_stores import (
+    VectorStores,
+    AsyncVectorStores,
+    VectorStoresWithRawResponse,
+    AsyncVectorStoresWithRawResponse,
+    VectorStoresWithStreamingResponse,
+    AsyncVectorStoresWithStreamingResponse,
+)
 
 __all__ = [
     "Completions",
@@ -152,6 +168,12 @@
     "AsyncFineTuningWithRawResponse",
     "FineTuningWithStreamingResponse",
     "AsyncFineTuningWithStreamingResponse",
+    "VectorStores",
+    "AsyncVectorStores",
+    "VectorStoresWithRawResponse",
+    "AsyncVectorStoresWithRawResponse",
+    "VectorStoresWithStreamingResponse",
+    "AsyncVectorStoresWithStreamingResponse",
     "Beta",
     "AsyncBeta",
     "BetaWithRawResponse",
@@ -170,4 +192,10 @@
     "AsyncUploadsWithRawResponse",
     "UploadsWithStreamingResponse",
     "AsyncUploadsWithStreamingResponse",
+    "Responses",
+    "AsyncResponses",
+    "ResponsesWithRawResponse",
+    "AsyncResponsesWithRawResponse",
+    "ResponsesWithStreamingResponse",
+    "AsyncResponsesWithStreamingResponse",
 ]

src/openai/resources/__init__.py

@@ -24,22 +24,8 @@
     AssistantsWithStreamingResponse,
     AsyncAssistantsWithStreamingResponse,
 )
-from .vector_stores import (
-    VectorStores,
-    AsyncVectorStores,
-    VectorStoresWithRawResponse,
-    AsyncVectorStoresWithRawResponse,
-    VectorStoresWithStreamingResponse,
-    AsyncVectorStoresWithStreamingResponse,
-)
 
 __all__ = [
-    "VectorStores",
-    "AsyncVectorStores",
-    "VectorStoresWithRawResponse",
-    "AsyncVectorStoresWithRawResponse",
-    "VectorStoresWithStreamingResponse",
-    "AsyncVectorStoresWithStreamingResponse",
     "Assistants",
     "AsyncAssistants",
     "AssistantsWithRawResponse",

src/openai/resources/beta/__init__.py

@@ -27,6 +27,7 @@
 from ...types.shared.chat_model import ChatModel
 from ...types.beta.assistant_deleted import AssistantDeleted
 from ...types.shared_params.metadata import Metadata
+from ...types.shared.reasoning_effort import ReasoningEffort
 from ...types.beta.assistant_tool_param import AssistantToolParam
 from ...types.beta.assistant_response_format_option_param import AssistantResponseFormatOptionParam
 
@@ -61,7 +62,7 @@ def create(
         instructions: Optional[str] | NotGiven = NOT_GIVEN,
         metadata: Optional[Metadata] | NotGiven = NOT_GIVEN,
         name: Optional[str] | NotGiven = NOT_GIVEN,
-        reasoning_effort: Optional[Literal["low", "medium", "high"]] | NotGiven = NOT_GIVEN,
+        reasoning_effort: Optional[ReasoningEffort] | NotGiven = NOT_GIVEN,
         response_format: Optional[AssistantResponseFormatOptionParam] | NotGiven = NOT_GIVEN,
         temperature: Optional[float] | NotGiven = NOT_GIVEN,
         tool_resources: Optional[assistant_create_params.ToolResources] | NotGiven = NOT_GIVEN,
@@ -98,7 +99,7 @@ def create(
 
           name: The name of the assistant. The maximum length is 256 characters.
 
-          reasoning_effort: **o1 and o3-mini models only**
+          reasoning_effort: **o-series models only**
 
               Constrains effort on reasoning for
               [reasoning models](https://platform.openai.com/docs/guides/reasoning). Currently
@@ -256,7 +257,7 @@ def update(
         ]
         | NotGiven = NOT_GIVEN,
         name: Optional[str] | NotGiven = NOT_GIVEN,
-        reasoning_effort: Optional[Literal["low", "medium", "high"]] | NotGiven = NOT_GIVEN,
+        reasoning_effort: Optional[ReasoningEffort] | NotGiven = NOT_GIVEN,
         response_format: Optional[AssistantResponseFormatOptionParam] | NotGiven = NOT_GIVEN,
         temperature: Optional[float] | NotGiven = NOT_GIVEN,
         tool_resources: Optional[assistant_update_params.ToolResources] | NotGiven = NOT_GIVEN,
@@ -294,7 +295,7 @@ def update(
 
           name: The name of the assistant. The maximum length is 256 characters.
 
-          reasoning_effort: **o1 and o3-mini models only**
+          reasoning_effort: **o-series models only**
 
               Constrains effort on reasoning for
               [reasoning models](https://platform.openai.com/docs/guides/reasoning). Currently
@@ -504,7 +505,7 @@ async def create(
         instructions: Optional[str] | NotGiven = NOT_GIVEN,
         metadata: Optional[Metadata] | NotGiven = NOT_GIVEN,
         name: Optional[str] | NotGiven = NOT_GIVEN,
-        reasoning_effort: Optional[Literal["low", "medium", "high"]] | NotGiven = NOT_GIVEN,
+        reasoning_effort: Optional[ReasoningEffort] | NotGiven = NOT_GIVEN,
         response_format: Optional[AssistantResponseFormatOptionParam] | NotGiven = NOT_GIVEN,
         temperature: Optional[float] | NotGiven = NOT_GIVEN,
         tool_resources: Optional[assistant_create_params.ToolResources] | NotGiven = NOT_GIVEN,
@@ -541,7 +542,7 @@ async def create(
 
           name: The name of the assistant. The maximum length is 256 characters.
 
-          reasoning_effort: **o1 and o3-mini models only**
+          reasoning_effort: **o-series models only**
 
               Constrains effort on reasoning for
               [reasoning models](https://platform.openai.com/docs/guides/reasoning). Currently
@@ -699,7 +700,7 @@ async def update(
         ]
         | NotGiven = NOT_GIVEN,
         name: Optional[str] | NotGiven = NOT_GIVEN,
-        reasoning_effort: Optional[Literal["low", "medium", "high"]] | NotGiven = NOT_GIVEN,
+        reasoning_effort: Optional[ReasoningEffort] | NotGiven = NOT_GIVEN,
         response_format: Optional[AssistantResponseFormatOptionParam] | NotGiven = NOT_GIVEN,
         temperature: Optional[float] | NotGiven = NOT_GIVEN,
         tool_resources: Optional[assistant_update_params.ToolResources] | NotGiven = NOT_GIVEN,
@@ -737,7 +738,7 @@ async def update(
 
           name: The name of the assistant. The maximum length is 256 characters.
 
-          reasoning_effort: **o1 and o3-mini models only**
+          reasoning_effort: **o-series models only**
 
               Constrains effort on reasoning for
               [reasoning models](https://platform.openai.com/docs/guides/reasoning). Currently

src/openai/resources/beta/assistants.py

@@ -28,14 +28,6 @@
     RealtimeWithStreamingResponse,
     AsyncRealtimeWithStreamingResponse,
 )
-from .vector_stores.vector_stores import (
-    VectorStores,
-    AsyncVectorStores,
-    VectorStoresWithRawResponse,
-    AsyncVectorStoresWithRawResponse,
-    VectorStoresWithStreamingResponse,
-    AsyncVectorStoresWithStreamingResponse,
-)
 
 __all__ = ["Beta", "AsyncBeta"]
 
@@ -45,10 +37,6 @@ class Beta(SyncAPIResource):
     def realtime(self) -> Realtime:
         return Realtime(self._client)
 
-    @cached_property
-    def vector_stores(self) -> VectorStores:
-        return VectorStores(self._client)
-
     @cached_property
     def assistants(self) -> Assistants:
         return Assistants(self._client)
@@ -82,10 +70,6 @@ class AsyncBeta(AsyncAPIResource):
     def realtime(self) -> AsyncRealtime:
         return AsyncRealtime(self._client)
 
-    @cached_property
-    def vector_stores(self) -> AsyncVectorStores:
-        return AsyncVectorStores(self._client)
-
     @cached_property
     def assistants(self) -> AsyncAssistants:
         return AsyncAssistants(self._client)
@@ -122,10 +106,6 @@ def __init__(self, beta: Beta) -> None:
     def realtime(self) -> RealtimeWithRawResponse:
         return RealtimeWithRawResponse(self._beta.realtime)
 
-    @cached_property
-    def vector_stores(self) -> VectorStoresWithRawResponse:
-        return VectorStoresWithRawResponse(self._beta.vector_stores)
-
     @cached_property
     def assistants(self) -> AssistantsWithRawResponse:
         return AssistantsWithRawResponse(self._beta.assistants)
@@ -143,10 +123,6 @@ def __init__(self, beta: AsyncBeta) -> None:
     def realtime(self) -> AsyncRealtimeWithRawResponse:
         return AsyncRealtimeWithRawResponse(self._beta.realtime)
 
-    @cached_property
-    def vector_stores(self) -> AsyncVectorStoresWithRawResponse:
-        return AsyncVectorStoresWithRawResponse(self._beta.vector_stores)
-
     @cached_property
     def assistants(self) -> AsyncAssistantsWithRawResponse:
         return AsyncAssistantsWithRawResponse(self._beta.assistants)
@@ -164,10 +140,6 @@ def __init__(self, beta: Beta) -> None:
     def realtime(self) -> RealtimeWithStreamingResponse:
         return RealtimeWithStreamingResponse(self._beta.realtime)
 
-    @cached_property
-    def vector_stores(self) -> VectorStoresWithStreamingResponse:
-        return VectorStoresWithStreamingResponse(self._beta.vector_stores)
-
     @cached_property
     def assistants(self) -> AssistantsWithStreamingResponse:
         return AssistantsWithStreamingResponse(self._beta.assistants)
@@ -185,10 +157,6 @@ def __init__(self, beta: AsyncBeta) -> None:
     def realtime(self) -> AsyncRealtimeWithStreamingResponse:
         return AsyncRealtimeWithStreamingResponse(self._beta.realtime)
 
-    @cached_property
-    def vector_stores(self) -> AsyncVectorStoresWithStreamingResponse:
-        return AsyncVectorStoresWithStreamingResponse(self._beta.vector_stores)
-
     @cached_property
     def assistants(self) -> AsyncAssistantsWithStreamingResponse:
         return AsyncAssistantsWithStreamingResponse(self._beta.assistants)