release: 1.107.2 (#2624)

* chore(api): Minor docs and type updates for realtime

* codegen metadata

* chore(tests): simplify `get_platform` test

`nest_asyncio` is archived and broken on some platforms so it's
not worth keeping in our test suite.

* release: 1.107.2

---------

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 @@
 {
-  ".": "1.107.1"
+  ".": "1.107.2"
 }

.stats.yml

@@ -1,4 +1,4 @@
 configured_endpoints: 118
-openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/openai%2Fopenai-16cb18bed32bae8c5840fb39a1bf664026cc40463ad0c487dcb0df1bd3d72db0.yml
-openapi_spec_hash: 4cb51b22f98dee1a90bc7add82d1d132
+openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/openai%2Fopenai-94b1e3cb0bdc616ff0c2f267c33dadd95f133b1f64e647aab6c64afb292b2793.yml
+openapi_spec_hash: 2395319ac9befd59b6536ae7f9564a05
 config_hash: 930dac3aa861344867e4ac84f037b5df

CHANGELOG.md

@@ -1,5 +1,14 @@
 # Changelog
 
+## 1.107.2 (2025-09-12)
+
+Full Changelog: [v1.107.1...v1.107.2](https://github.com/openai/openai-python/compare/v1.107.1...v1.107.2)
+
+### Chores
+
+* **api:** Minor docs and type updates for realtime ([ab6a10d](https://github.com/openai/openai-python/commit/ab6a10da4ed7e6386695b6f5f29149d4870f85c9))
+* **tests:** simplify `get_platform` test ([01f03e0](https://github.com/openai/openai-python/commit/01f03e0ad1f9ab3f2ed8b7c13d652263c6d06378))
+
 ## 1.107.1 (2025-09-10)
 
 Full Changelog: [v1.107.0...v1.107.1](https://github.com/openai/openai-python/compare/v1.107.0...v1.107.1)

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "openai"
-version = "1.107.1"
+version = "1.107.2"
 description = "The official Python library for the openai API"
 dynamic = ["readme"]
 license = "Apache-2.0"

requirements-dev.lock

@@ -70,7 +70,7 @@ filelock==3.12.4
 frozenlist==1.7.0
     # via aiohttp
     # via aiosignal
-griffe==1.14.0
+griffe==1.13.0
 h11==0.16.0
     # via httpcore
 httpcore==1.0.9
@@ -108,7 +108,6 @@ multidict==6.5.0
 mypy==1.14.1
 mypy-extensions==1.0.0
     # via mypy
-nest-asyncio==1.6.0
 nodeenv==1.8.0
     # via pyright
 nox==2023.4.22

src/openai/_version.py

@@ -1,4 +1,4 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
 __title__ = "openai"
-__version__ = "1.107.1"  # x-release-please-version
+__version__ = "1.107.2"  # x-release-please-version

src/openai/resources/responses/responses.py

@@ -288,10 +288,10 @@ def create(
 
           truncation: The truncation strategy to use for the model response.
 
-              - `auto`: If the context of this response and previous ones exceeds the model's
-                context window size, the model will truncate the response to fit the context
-                window by dropping input items in the middle of the conversation.
-              - `disabled` (default): If a model response will exceed the context window size
+              - `auto`: If the input to this Response exceeds the model's context window size,
+                the model will truncate the response to fit the context window by dropping
+                items from the beginning of the conversation.
+              - `disabled` (default): If the input size will exceed the context window size
                 for a model, the request will fail with a 400 error.
 
           user: This field is being replaced by `safety_identifier` and `prompt_cache_key`. Use
@@ -527,10 +527,10 @@ def create(
 
           truncation: The truncation strategy to use for the model response.
 
-              - `auto`: If the context of this response and previous ones exceeds the model's
-                context window size, the model will truncate the response to fit the context
-                window by dropping input items in the middle of the conversation.
-              - `disabled` (default): If a model response will exceed the context window size
+              - `auto`: If the input to this Response exceeds the model's context window size,
+                the model will truncate the response to fit the context window by dropping
+                items from the beginning of the conversation.
+              - `disabled` (default): If the input size will exceed the context window size
                 for a model, the request will fail with a 400 error.
 
           user: This field is being replaced by `safety_identifier` and `prompt_cache_key`. Use
@@ -766,10 +766,10 @@ def create(
 
           truncation: The truncation strategy to use for the model response.
 
-              - `auto`: If the context of this response and previous ones exceeds the model's
-                context window size, the model will truncate the response to fit the context
-                window by dropping input items in the middle of the conversation.
-              - `disabled` (default): If a model response will exceed the context window size
+              - `auto`: If the input to this Response exceeds the model's context window size,
+                the model will truncate the response to fit the context window by dropping
+                items from the beginning of the conversation.
+              - `disabled` (default): If the input size will exceed the context window size
                 for a model, the request will fail with a 400 error.
 
           user: This field is being replaced by `safety_identifier` and `prompt_cache_key`. Use
@@ -1719,10 +1719,10 @@ async def create(
 
           truncation: The truncation strategy to use for the model response.
 
-              - `auto`: If the context of this response and previous ones exceeds the model's
-                context window size, the model will truncate the response to fit the context
-                window by dropping input items in the middle of the conversation.
-              - `disabled` (default): If a model response will exceed the context window size
+              - `auto`: If the input to this Response exceeds the model's context window size,
+                the model will truncate the response to fit the context window by dropping
+                items from the beginning of the conversation.
+              - `disabled` (default): If the input size will exceed the context window size
                 for a model, the request will fail with a 400 error.
 
           user: This field is being replaced by `safety_identifier` and `prompt_cache_key`. Use
@@ -1958,10 +1958,10 @@ async def create(
 
           truncation: The truncation strategy to use for the model response.
 
-              - `auto`: If the context of this response and previous ones exceeds the model's
-                context window size, the model will truncate the response to fit the context
-                window by dropping input items in the middle of the conversation.
-              - `disabled` (default): If a model response will exceed the context window size
+              - `auto`: If the input to this Response exceeds the model's context window size,
+                the model will truncate the response to fit the context window by dropping
+                items from the beginning of the conversation.
+              - `disabled` (default): If the input size will exceed the context window size
                 for a model, the request will fail with a 400 error.
 
           user: This field is being replaced by `safety_identifier` and `prompt_cache_key`. Use
@@ -2197,10 +2197,10 @@ async def create(
 
           truncation: The truncation strategy to use for the model response.
 
-              - `auto`: If the context of this response and previous ones exceeds the model's
-                context window size, the model will truncate the response to fit the context
-                window by dropping input items in the middle of the conversation.
-              - `disabled` (default): If a model response will exceed the context window size
+              - `auto`: If the input to this Response exceeds the model's context window size,
+                the model will truncate the response to fit the context window by dropping
+                items from the beginning of the conversation.
+              - `disabled` (default): If the input size will exceed the context window size
                 for a model, the request will fail with a 400 error.
 
           user: This field is being replaced by `safety_identifier` and `prompt_cache_key`. Use

src/openai/types/realtime/input_audio_buffer_timeout_triggered.py

@@ -9,10 +9,16 @@
 
 class InputAudioBufferTimeoutTriggered(BaseModel):
     audio_end_ms: int
-    """Millisecond offset where speech ended within the buffered audio."""
+    """
+    Millisecond offset of audio written to the input audio buffer at the time the
+    timeout was triggered.
+    """
 
     audio_start_ms: int
-    """Millisecond offset where speech started within the buffered audio."""
+    """
+    Millisecond offset of audio written to the input audio buffer that was after the
+    playback time of the last model response.
+    """
 
     event_id: str
     """The unique ID of the server event."""

src/openai/types/realtime/realtime_audio_config_input.py

@@ -49,8 +49,11 @@ class RealtimeAudioConfigInput(BaseModel):
     """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.
+    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 conjunction
     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

src/openai/types/realtime/realtime_audio_config_input_param.py

@@ -2,6 +2,7 @@
 
 from __future__ import annotations
 
+from typing import Optional
 from typing_extensions import TypedDict
 
 from .noise_reduction_type import NoiseReductionType
@@ -46,12 +47,15 @@ class RealtimeAudioConfigInputParam(TypedDict, total=False):
     transcription, these offer additional guidance to the transcription service.
     """
 
-    turn_detection: RealtimeAudioInputTurnDetectionParam
+    turn_detection: Optional[RealtimeAudioInputTurnDetectionParam]
     """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.
+    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 conjunction
     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