Merge branch 'main' into dev/valenradovich/parse-return-validationerror

Author: valenradovich

.github/workflows/ci.yml

@@ -6,6 +6,7 @@ on:
   pull_request:
     branches:
       - main
+      - next
 
 jobs:
   lint:

.github/workflows/create-releases.yml

@@ -0,0 +1,39 @@
+name: Create releases
+on:
+  schedule:
+    - cron: '0 5 * * *' # every day at 5am UTC
+  push:
+    branches:
+      - main
+
+jobs:
+  release:
+    name: release
+    if: github.ref == 'refs/heads/main' && github.repository == 'openai/openai-python'
+    runs-on: ubuntu-latest
+    environment: publish
+
+    steps:
+      - uses: actions/checkout@v4
+
+      - uses: stainless-api/trigger-release-please@v1
+        id: release
+        with:
+          repo: ${{ github.event.repository.full_name }}
+          stainless-api-key: ${{ secrets.STAINLESS_API_KEY }}
+
+      - name: Install Rye
+        if: ${{ steps.release.outputs.releases_created }}
+        run: |
+          curl -sSf https://rye.astral.sh/get | bash
+          echo "$HOME/.rye/shims" >> $GITHUB_PATH
+        env:
+          RYE_VERSION: '0.44.0'
+          RYE_INSTALL_OPTION: '--yes'
+
+      - name: Publish to PyPI
+        if: ${{ steps.release.outputs.releases_created }}
+        run: |
+          bash ./bin/publish-pypi
+        env:
+          PYPI_TOKEN: ${{ secrets.OPENAI_PYPI_TOKEN || secrets.PYPI_TOKEN }}

.gitignore

@@ -14,3 +14,7 @@ dist
 .envrc
 codegen.log
 Brewfile.lock.json
+
+.DS_Store
+
+examples/*.mp3

.release-please-manifest.json

@@ -1,3 +1,3 @@
 {
-  ".": "1.66.4"
+  ".": "1.68.2"
 }

.stats.yml

@@ -1,2 +1,2 @@
-configured_endpoints: 81
-openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/openai%2Fopenai-c8579861bc21d4d2155a5b9e8e7d54faee8083730673c4d32cbbe573d7fb4116.yml
+configured_endpoints: 82
+openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/openai%2Fopenai-5ad6884898c07591750dde560118baf7074a59aecd1f367f930c5e42b04e848a.yml

CHANGELOG.md

@@ -1,5 +1,59 @@
 # Changelog
 
+## 1.68.2 (2025-03-21)
+
+Full Changelog: [v1.68.1...v1.68.2](https://github.com/openai/openai-python/compare/v1.68.1...v1.68.2)
+
+### Refactors
+
+* **package:** rename audio extra to voice_helpers ([2dd6cb8](https://github.com/openai/openai-python/commit/2dd6cb87489fe12c5e45128f44d985c3f49aba1d))
+
+## 1.68.1 (2025-03-21)
+
+Full Changelog: [v1.68.0...v1.68.1](https://github.com/openai/openai-python/compare/v1.68.0...v1.68.1)
+
+### Bug Fixes
+
+* **client:** remove duplicate types ([#2235](https://github.com/openai/openai-python/issues/2235)) ([063f7d0](https://github.com/openai/openai-python/commit/063f7d0684c350ca9d766e2cb150233a22a623c8))
+* **helpers/audio:** remove duplicative module ([f253d04](https://github.com/openai/openai-python/commit/f253d0415145f2c4904ea2e7b389d31d94e45a54))
+* **package:** make sounddevice and numpy optional dependencies ([8b04453](https://github.com/openai/openai-python/commit/8b04453f0483736c13f0209a9f8f3618bc0e86c9))
+
+
+### Chores
+
+* **ci:** run workflows on next too ([67f89d4](https://github.com/openai/openai-python/commit/67f89d478aab780d1481c9bf6682c6633e431137))
+
+## 1.68.0 (2025-03-20)
+
+Full Changelog: [v1.67.0...v1.68.0](https://github.com/openai/openai-python/compare/v1.67.0...v1.68.0)
+
+### Features
+
+* add audio helpers ([423655c](https://github.com/openai/openai-python/commit/423655ca9077cfd258f1e52f6eb386fc8307fa5f))
+* **api:** new models for TTS, STT, + new audio features for Realtime ([#2232](https://github.com/openai/openai-python/issues/2232)) ([ab5192d](https://github.com/openai/openai-python/commit/ab5192d0a7b417ade622ec94dd48f86beb90692c))
+
+## 1.67.0 (2025-03-19)
+
+Full Changelog: [v1.66.5...v1.67.0](https://github.com/openai/openai-python/compare/v1.66.5...v1.67.0)
+
+### Features
+
+* **api:** o1-pro now available through the API ([#2228](https://github.com/openai/openai-python/issues/2228)) ([40a19d8](https://github.com/openai/openai-python/commit/40a19d8592c1767d6318230fc93e37c360d1bcd1))
+
+## 1.66.5 (2025-03-18)
+
+Full Changelog: [v1.66.4...v1.66.5](https://github.com/openai/openai-python/compare/v1.66.4...v1.66.5)
+
+### Bug Fixes
+
+* **types:** improve responses type names ([#2224](https://github.com/openai/openai-python/issues/2224)) ([5f7beb8](https://github.com/openai/openai-python/commit/5f7beb873af5ccef2551f34ab3ef098e099ce9c6))
+
+
+### Chores
+
+* **internal:** add back releases workflow ([c71d4c9](https://github.com/openai/openai-python/commit/c71d4c918eab3532b36ea944b0c4069db6ac2d38))
+* **internal:** codegen related update ([#2222](https://github.com/openai/openai-python/issues/2222)) ([f570d91](https://github.com/openai/openai-python/commit/f570d914a16cb5092533e32dfd863027d378c0b5))
+
 ## 1.66.4 (2025-03-17)
 
 Full Changelog: [v1.66.3...v1.66.4](https://github.com/openai/openai-python/compare/v1.66.3...v1.66.4)

api.md

@@ -2,6 +2,7 @@
 
 ```python
 from openai.types import (
+    AllModels,
     ChatModel,
     ComparisonFilter,
     CompoundFilter,
@@ -14,6 +15,7 @@ from openai.types import (
     ResponseFormatJSONObject,
     ResponseFormatJSONSchema,
     ResponseFormatText,
+    ResponsesModel,
 )
 ```
 
@@ -149,7 +151,11 @@ Types:
 ```python
 from openai.types.audio import (
     Transcription,
+    TranscriptionInclude,
     TranscriptionSegment,
+    TranscriptionStreamEvent,
+    TranscriptionTextDeltaEvent,
+    TranscriptionTextDoneEvent,
     TranscriptionVerbose,
     TranscriptionWord,
     TranscriptionCreateResponse,
@@ -336,7 +342,9 @@ from openai.types.beta.realtime import (
     ConversationItemDeleteEvent,
     ConversationItemDeletedEvent,
     ConversationItemInputAudioTranscriptionCompletedEvent,
+    ConversationItemInputAudioTranscriptionDeltaEvent,
     ConversationItemInputAudioTranscriptionFailedEvent,
+    ConversationItemRetrieveEvent,
     ConversationItemTruncateEvent,
     ConversationItemTruncatedEvent,
     ConversationItemWithReference,
@@ -373,6 +381,8 @@ from openai.types.beta.realtime import (
     SessionCreatedEvent,
     SessionUpdateEvent,
     SessionUpdatedEvent,
+    TranscriptionSessionUpdate,
+    TranscriptionSessionUpdatedEvent,
 )
 ```
 
@@ -388,6 +398,18 @@ Methods:
 
 - <code title="post /realtime/sessions">client.beta.realtime.sessions.<a href="./src/openai/resources/beta/realtime/sessions.py">create</a>(\*\*<a href="src/openai/types/beta/realtime/session_create_params.py">params</a>) -> <a href="./src/openai/types/beta/realtime/session_create_response.py">SessionCreateResponse</a></code>
 
+### TranscriptionSessions
+
+Types:
+
+```python
+from openai.types.beta.realtime import TranscriptionSession
+```
+
+Methods:
+
+- <code title="post /realtime/transcription_sessions">client.beta.realtime.transcription_sessions.<a href="./src/openai/resources/beta/realtime/transcription_sessions.py">create</a>(\*\*<a href="src/openai/types/beta/realtime/transcription_session_create_params.py">params</a>) -> <a href="./src/openai/types/beta/realtime/transcription_session.py">TranscriptionSession</a></code>
+
 ## Assistants
 
 Types:
@@ -605,6 +627,8 @@ from openai.types.responses import (
     ResponseCodeInterpreterToolCall,
     ResponseCompletedEvent,
     ResponseComputerToolCall,
+    ResponseComputerToolCallOutputItem,
+    ResponseComputerToolCallOutputScreenshot,
     ResponseContent,
     ResponseContentPartAddedEvent,
     ResponseContentPartDoneEvent,
@@ -621,6 +645,8 @@ from openai.types.responses import (
     ResponseFunctionCallArgumentsDeltaEvent,
     ResponseFunctionCallArgumentsDoneEvent,
     ResponseFunctionToolCall,
+    ResponseFunctionToolCallItem,
+    ResponseFunctionToolCallOutputItem,
     ResponseFunctionWebSearch,
     ResponseInProgressEvent,
     ResponseIncludable,
@@ -632,7 +658,9 @@ from openai.types.responses import (
     ResponseInputImage,
     ResponseInputItem,
     ResponseInputMessageContentList,
+    ResponseInputMessageItem,
     ResponseInputText,
+    ResponseItem,
     ResponseOutputAudio,
     ResponseOutputItem,
     ResponseOutputItemAddedEvent,
@@ -677,4 +705,4 @@ from openai.types.responses import ResponseItemList
 
 Methods:
 
-- <code title="get /responses/{response_id}/input_items">client.responses.input_items.<a href="./src/openai/resources/responses/input_items.py">list</a>(response_id, \*\*<a href="src/openai/types/responses/input_item_list_params.py">params</a>) -> SyncCursorPage[Data]</code>
+- <code title="get /responses/{response_id}/input_items">client.responses.input_items.<a href="./src/openai/resources/responses/input_items.py">list</a>(response_id, \*\*<a href="src/openai/types/responses/input_item_list_params.py">params</a>) -> <a href="./src/openai/types/responses/response_item.py">SyncCursorPage[ResponseItem]</a></code>

examples/audio.py

@@ -1,6 +1,5 @@
 #!/usr/bin/env rye run python
 
-import time
 from pathlib import Path
 
 from openai import OpenAI
@@ -12,8 +11,6 @@
 
 
 def main() -> None:
-    stream_to_speakers()
-
     # Create text-to-speech audio file
     with openai.audio.speech.with_streaming_response.create(
         model="tts-1",
@@ -37,28 +34,5 @@ def main() -> None:
     print(translation.text)
 
 
-def stream_to_speakers() -> None:
-    import pyaudio
-
-    player_stream = pyaudio.PyAudio().open(format=pyaudio.paInt16, channels=1, rate=24000, output=True)
-
-    start_time = time.time()
-
-    with openai.audio.speech.with_streaming_response.create(
-        model="tts-1",
-        voice="alloy",
-        response_format="pcm",  # similar to WAV, but without a header chunk at the start.
-        input="""I see skies of blue and clouds of white
-                The bright blessed days, the dark sacred nights
-                And I think to myself
-                What a wonderful world""",
-    ) as response:
-        print(f"Time to first byte: {int((time.time() - start_time) * 1000)}ms")
-        for chunk in response.iter_bytes(chunk_size=1024):
-            player_stream.write(chunk)
-
-    print(f"Done in {int((time.time() - start_time) * 1000)}ms.")
-
-
 if __name__ == "__main__":
     main()

examples/speech_to_text.py

@@ -0,0 +1,25 @@
+#!/usr/bin/env rye run python
+
+import asyncio
+
+from openai import AsyncOpenAI
+from openai.helpers import Microphone
+
+# gets OPENAI_API_KEY from your environment variables
+openai = AsyncOpenAI()
+
+
+async def main() -> None:
+    print("Recording for the next 10 seconds...")
+    recording = await Microphone(timeout=10).record()
+    print("Recording complete")
+    transcription = await openai.audio.transcriptions.create(
+        model="whisper-1",
+        file=recording,
+    )
+
+    print(transcription.text)
+
+
+if __name__ == "__main__":
+    asyncio.run(main())

examples/text_to_speech.py

@@ -0,0 +1,31 @@
+#!/usr/bin/env rye run python
+
+import time
+import asyncio
+
+from openai import AsyncOpenAI
+from openai.helpers import LocalAudioPlayer
+
+# gets OPENAI_API_KEY from your environment variables
+openai = AsyncOpenAI()
+
+
+async def main() -> None:
+    start_time = time.time()
+
+    async with openai.audio.speech.with_streaming_response.create(
+        model="tts-1",
+        voice="alloy",
+        response_format="pcm",  # similar to WAV, but without a header chunk at the start.
+        input="""I see skies of blue and clouds of white
+                The bright blessed days, the dark sacred nights
+                And I think to myself
+                What a wonderful world""",
+    ) as response:
+        print(f"Time to first byte: {int((time.time() - start_time) * 1000)}ms")
+        await LocalAudioPlayer().play(response)
+        print(f"Time to play: {int((time.time() - start_time) * 1000)}ms")
+
+
+if __name__ == "__main__":
+    asyncio.run(main())

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "openai"
-version = "1.66.4"
+version = "1.68.2"
 description = "The official Python library for the openai API"
 dynamic = ["readme"]
 license = "Apache-2.0"
@@ -45,6 +45,7 @@ openai = "openai.cli:main"
 [project.optional-dependencies]
 realtime = ["websockets >= 13, < 15"]
 datalib = ["numpy >= 1", "pandas >= 1.2.3", "pandas-stubs >= 1.1.0.11"]
+voice_helpers = ["sounddevice>=0.5.1", "numpy>=2.0.2"]
 
 [tool.rye]
 managed = true

requirements-dev.lock

@@ -33,6 +33,7 @@ certifi==2023.7.22
     # via requests
 cffi==1.16.0
     # via cryptography
+    # via sounddevice
 charset-normalizer==3.3.2
     # via requests
 click==8.1.7
@@ -92,7 +93,7 @@ nest-asyncio==1.6.0
 nodeenv==1.8.0
     # via pyright
 nox==2023.4.22
-numpy==1.26.3
+numpy==2.0.2
     # via openai
     # via pandas
     # via pandas-stubs
@@ -102,7 +103,7 @@ packaging==23.2
     # via black
     # via nox
     # via pytest
-pandas==2.1.4
+pandas==2.2.3
     # via openai
 pandas-stubs==2.1.4.231227
     # via openai
@@ -154,6 +155,8 @@ sniffio==1.3.0
     # via trio
 sortedcontainers==2.4.0
     # via trio
+sounddevice==0.5.1
+    # via openai
 time-machine==2.9.0
 toml==0.10.2
     # via inline-snapshot