release: 2.10.0 (#2767)

* chore(internal): update docstring

* fix(types): allow pyright to infer TypedDict types within SequenceNotStr

* chore: add missing docstrings

* feat(api): make model required for the responses/compact endpoint

* release: 2.10.0

---------

Co-authored-by: stainless-app[bot] <142633134+stainless-app[bot]@users.noreply.github.com>

Author: stainless-app[bot]

.release-please-manifest.json

@@ -1,3 +1,3 @@
 {
-  ".": "2.9.0"
+  ".": "2.10.0"
 }

.stats.yml

@@ -1,4 +1,4 @@
 configured_endpoints: 137
-openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/openai%2Fopenai-fe8a79e6fd407e6c9afec60971f03076b65f711ccd6ea16457933b0e24fb1f6d.yml
-openapi_spec_hash: 38c0a73f4e08843732c5f8002a809104
-config_hash: 2c350086d87a4b4532077363087840e7
+openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/openai%2Fopenai-88d85ff87ad8983262af2b729762a6e05fd509468bb691529bc2f81e4ce27c69.yml
+openapi_spec_hash: 46a55acbccd0147534017b92c1f4dd99
+config_hash: 141b101c9f13b90e21af74e1686f1f41

CHANGELOG.md

@@ -1,5 +1,24 @@
 # Changelog
 
+## 2.10.0 (2025-12-10)
+
+Full Changelog: [v2.9.0...v2.10.0](https://github.com/openai/openai-python/compare/v2.9.0...v2.10.0)
+
+### Features
+
+* **api:** make model required for the responses/compact endpoint ([a12936b](https://github.com/openai/openai-python/commit/a12936b18cf19009d4e6d586c9b1958359636dbe))
+
+
+### Bug Fixes
+
+* **types:** allow pyright to infer TypedDict types within SequenceNotStr ([8f0d230](https://github.com/openai/openai-python/commit/8f0d23066c1edc38a6e9858b054dceaf92ae001b))
+
+
+### Chores
+
+* add missing docstrings ([f20a9a1](https://github.com/openai/openai-python/commit/f20a9a18a421ba69622c77ab539509d218e774eb))
+* **internal:** update docstring ([9a993f2](https://github.com/openai/openai-python/commit/9a993f2261b6524aa30b955e006c7ea89f086968))
+
 ## 2.9.0 (2025-12-04)
 
 Full Changelog: [v2.8.1...v2.9.0](https://github.com/openai/openai-python/compare/v2.8.1...v2.9.0)

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "openai"
-version = "2.9.0"
+version = "2.10.0"
 description = "The official Python library for the openai API"
 dynamic = ["readme"]
 license = "Apache-2.0"

src/openai/_types.py

@@ -247,6 +247,9 @@ class HttpxSendArgs(TypedDict, total=False):
 if TYPE_CHECKING:
     # This works because str.__contains__ does not accept object (either in typeshed or at runtime)
     # https://github.com/hauntsaninja/useful_types/blob/5e9710f3875107d068e7679fd7fec9cfab0eff3b/useful_types/__init__.py#L285
+    #
+    # Note: index() and count() methods are intentionally omitted to allow pyright to properly
+    # infer TypedDict types when dict literals are used in lists assigned to SequenceNotStr.
     class SequenceNotStr(Protocol[_T_co]):
         @overload
         def __getitem__(self, index: SupportsIndex, /) -> _T_co: ...
@@ -255,8 +258,6 @@ def __getitem__(self, index: slice, /) -> Sequence[_T_co]: ...
         def __contains__(self, value: object, /) -> bool: ...
         def __len__(self) -> int: ...
         def __iter__(self) -> Iterator[_T_co]: ...
-        def index(self, value: Any, start: int = 0, stop: int = ..., /) -> int: ...
-        def count(self, value: Any, /) -> int: ...
         def __reversed__(self) -> Iterator[_T_co]: ...
 else:
     # just point this to a normal `Sequence` at runtime to avoid having to special case

src/openai/_version.py

@@ -1,4 +1,4 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
 __title__ = "openai"
-__version__ = "2.9.0"  # x-release-please-version
+__version__ = "2.10.0"  # x-release-please-version

src/openai/resources/realtime/realtime.py

@@ -232,7 +232,7 @@ def calls(self) -> AsyncCallsWithStreamingResponse:
 
 
 class AsyncRealtimeConnection:
-    """Represents a live websocket connection to the Realtime API"""
+    """Represents a live WebSocket connection to the Realtime API"""
 
     session: AsyncRealtimeSessionResource
     response: AsyncRealtimeResponseResource
@@ -421,7 +421,7 @@ async def __aexit__(
 
 
 class RealtimeConnection:
-    """Represents a live websocket connection to the Realtime API"""
+    """Represents a live WebSocket connection to the Realtime API"""
 
     session: RealtimeSessionResource
     response: RealtimeResponseResource

src/openai/resources/responses/responses.py

@@ -1526,8 +1526,6 @@ def cancel(
     def compact(
         self,
         *,
-        input: Union[str, Iterable[ResponseInputItemParam], None] | Omit = omit,
-        instructions: Optional[str] | Omit = omit,
         model: Union[
             Literal[
                 "gpt-5.1",
@@ -1614,8 +1612,9 @@ def compact(
             ],
             str,
             None,
-        ]
-        | Omit = omit,
+        ],
+        input: Union[str, Iterable[ResponseInputItemParam], None] | Omit = omit,
+        instructions: Optional[str] | Omit = omit,
         previous_response_id: Optional[str] | Omit = omit,
         # 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.
@@ -1628,19 +1627,19 @@ def compact(
         Compact conversation
 
         Args:
+          model: Model ID used to generate the response, like `gpt-5` or `o3`. OpenAI offers a
+              wide range of models with different capabilities, performance characteristics,
+              and price points. Refer to the
+              [model guide](https://platform.openai.com/docs/models) to browse and compare
+              available models.
+
           input: Text, image, or file inputs to the model, used to generate a response
 
           instructions: A system (or developer) message inserted into the model's context. When used
               along with `previous_response_id`, the instructions from a previous response
               will not be carried over to the next response. This makes it simple to swap out
               system (or developer) messages in new responses.
 
-          model: Model ID used to generate the response, like `gpt-5` or `o3`. OpenAI offers a
-              wide range of models with different capabilities, performance characteristics,
-              and price points. Refer to the
-              [model guide](https://platform.openai.com/docs/models) to browse and compare
-              available models.
-
           previous_response_id: The unique ID of the previous response to the model. Use this to create
               multi-turn conversations. Learn more about
               [conversation state](https://platform.openai.com/docs/guides/conversation-state).
@@ -1658,9 +1657,9 @@ def compact(
             "/responses/compact",
             body=maybe_transform(
                 {
+                    "model": model,
                     "input": input,
                     "instructions": instructions,
-                    "model": model,
                     "previous_response_id": previous_response_id,
                 },
                 response_compact_params.ResponseCompactParams,
@@ -3140,8 +3139,6 @@ async def cancel(
     async def compact(
         self,
         *,
-        input: Union[str, Iterable[ResponseInputItemParam], None] | Omit = omit,
-        instructions: Optional[str] | Omit = omit,
         model: Union[
             Literal[
                 "gpt-5.1",
@@ -3228,8 +3225,9 @@ async def compact(
             ],
             str,
             None,
-        ]
-        | Omit = omit,
+        ],
+        input: Union[str, Iterable[ResponseInputItemParam], None] | Omit = omit,
+        instructions: Optional[str] | Omit = omit,
         previous_response_id: Optional[str] | Omit = omit,
         # 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.
@@ -3242,19 +3240,19 @@ async def compact(
         Compact conversation
 
         Args:
+          model: Model ID used to generate the response, like `gpt-5` or `o3`. OpenAI offers a
+              wide range of models with different capabilities, performance characteristics,
+              and price points. Refer to the
+              [model guide](https://platform.openai.com/docs/models) to browse and compare
+              available models.
+
           input: Text, image, or file inputs to the model, used to generate a response
 
           instructions: A system (or developer) message inserted into the model's context. When used
               along with `previous_response_id`, the instructions from a previous response
               will not be carried over to the next response. This makes it simple to swap out
               system (or developer) messages in new responses.
 
-          model: Model ID used to generate the response, like `gpt-5` or `o3`. OpenAI offers a
-              wide range of models with different capabilities, performance characteristics,
-              and price points. Refer to the
-              [model guide](https://platform.openai.com/docs/models) to browse and compare
-              available models.
-
           previous_response_id: The unique ID of the previous response to the model. Use this to create
               multi-turn conversations. Learn more about
               [conversation state](https://platform.openai.com/docs/guides/conversation-state).
@@ -3272,9 +3270,9 @@ async def compact(
             "/responses/compact",
             body=await async_maybe_transform(
                 {
+                    "model": model,
                     "input": input,
                     "instructions": instructions,
-                    "model": model,
                     "previous_response_id": previous_response_id,
                 },
                 response_compact_params.ResponseCompactParams,

src/openai/types/audio/transcription.py

@@ -21,6 +21,8 @@ class Logprob(BaseModel):
 
 
 class UsageTokensInputTokenDetails(BaseModel):
+    """Details about the input tokens billed for this request."""
+
     audio_tokens: Optional[int] = None
     """Number of audio tokens billed for this request."""
 
@@ -29,6 +31,8 @@ class UsageTokensInputTokenDetails(BaseModel):
 
 
 class UsageTokens(BaseModel):
+    """Usage statistics for models billed by token usage."""
+
     input_tokens: int
     """Number of input tokens billed for this request."""
 
@@ -46,6 +50,8 @@ class UsageTokens(BaseModel):
 
 
 class UsageDuration(BaseModel):
+    """Usage statistics for models billed by audio input duration."""
+
     seconds: float
     """Duration of the input audio in seconds."""
 
@@ -57,6 +63,10 @@ class UsageDuration(BaseModel):
 
 
 class Transcription(BaseModel):
+    """
+    Represents a transcription response returned by model, based on the provided input.
+    """
+
     text: str
     """The transcribed text."""
 

src/openai/types/audio/transcription_diarized.py

@@ -11,6 +11,8 @@
 
 
 class UsageTokensInputTokenDetails(BaseModel):
+    """Details about the input tokens billed for this request."""
+
     audio_tokens: Optional[int] = None
     """Number of audio tokens billed for this request."""
 
@@ -19,6 +21,8 @@ class UsageTokensInputTokenDetails(BaseModel):
 
 
 class UsageTokens(BaseModel):
+    """Usage statistics for models billed by token usage."""
+
     input_tokens: int
     """Number of input tokens billed for this request."""
 
@@ -36,6 +40,8 @@ class UsageTokens(BaseModel):
 
 
 class UsageDuration(BaseModel):
+    """Usage statistics for models billed by audio input duration."""
+
     seconds: float
     """Duration of the input audio in seconds."""
 
@@ -47,6 +53,10 @@ class UsageDuration(BaseModel):
 
 
 class TranscriptionDiarized(BaseModel):
+    """
+    Represents a diarized transcription response returned by the model, including the combined transcript and speaker-segment annotations.
+    """
+
     duration: float
     """Duration of the input audio in seconds."""