NewYour coding agent can read the release notes before it upgrades.Set up the MCP server →
PyPI · #2567 most downloaded on PyPI
Agent Framework plugin for services from OpenAI
Last release 3 days ago
01 Oct 2026
Ships fairly regularly
a new release about every 8 days
Rarely documented
notes for 6 of the last 60 stable releases
Nothing withdrawn
no release was ever pulled
3 years old
184 releases · first in 2023
openai: default to use_realtime=False - #1783 (@davidzhao)
openai: default to use_realtime=False - #1783 (@davidzhao)
fix(openai): pass NotGiven to OpenAI when instructions are omitted - #1834 (@davidzhao)
fix: openai stt error when using detect language - #1755 (@jayeshp19)
One column per month.
expose turn_detection options with openai STT - #1726 (@davidzhao)
expose turn_detection options with openai STT - #1726 (@davidzhao)
feat(OpenAI STT): add support for semantic_vad - #1707 (@chasemcdo)
support for streaming STT, new STT/TTS models - #1701 (@davidzhao)
Make azure and openai take a timeout optionally. Also update the default timeout for Azure OpenAI to 5s from 10 minutes. - #1674 (@martin-purplefish)
Support more input transcription parameters for openai realtime - #1637 (@adambenali)
Support more input transcription parameters for openai realtime - #1637 (@adambenali)
Add string type support to model parameter - #1657 (@jayeshp19)
version bump to 0.11.1 - #1640 (@davidzhao)
openai tts: switch to using Opus encoding - #1494 (@davidzhao)
fix: [openai] only send params when set - #1474 (@jayeshp19)
fix: [openai] only send params when set - #1474 (@jayeshp19)
fix response create for openai realtime model - #1469 (@longcw)
Added an additional field in LLM capabilities class to check if model providers support function call history within chat context without needing func
gemini-realtime: fix input audio sample rate - #1411 (@jayeshp19)
add generate_reply api for multimodal agent - #1359 (@longcw)
support disabling server VAD for OpenAI realtime model - #1347 (@longcw)
fix: revert from weakset to list in multimodal for maintaining sessions - #1326 (@jayeshp19)
improved handling of LLM errors, do not retry if already began - #1298 (@davidzhao)
improved handling of LLM errors, do not retry if already began - #1298 (@davidzhao)
make multimodal class generic and support gemini live api - #1240 (@jayeshp19)
fix unknown metadata & store fields on OpenAI-like API - #1276 (@theomonnom)
metadata & store fields on OpenAI-like API - #1276 (@theomonnom)Moved create_ai_function_info to function_context.py for better reusability and reduce repetation - #1260 (@jayeshp19)
Moved create_ai_function_info to function_context.py for better reusability and reduce repetation - #1260 (@jayeshp19)
add on_duplicate option for multimodal agent response create - #1204 (@longcw)
Add support for OpenAI's "detail" parameter to ChatImage - #1213 (@bcherry)
Add support for data URLs on ChatImage in the Anthropic plugin.
filter out empty message for set chat ctx in realtime model - #1245 (@longcw)
fix: correctly parse function argument types - #1221 (@jayeshp19)
add session_updated event for RealtimeSession - #1253 (@longcw)
added llama 3.3 70b to model definitions - #1233 (@davidzhao)
update default realtime model to gpt-4o-realtime-preview-2024-12-17 - #1250 (@davidzhao)
Fix center_aspect_fit bug, add scale_aspect_fit and scale_aspect_fill resizing options. - #1222 (@bcherry)
Make scale_aspect_fit the new default resizing option for video frames.
add google/gemini-2.0-flash-exp as default model for vertex - #1214 (@jayeshp19)
add google/gemini-2.0-flash-exp as default model for vertex - #1214 (@jayeshp19)
fix: return structured output from func calls - #1187 (@jayeshp19)
Handle optional func args in tool calls when set to None - #1211 (@jayeshp19)
fix: openai llm retries - #1196 (@theomonnom)
Improvements to end of turn plugin, ensure STT language settings. - #1195 (@davidzhao)
fix: Handle optional func args in tool calls when set to None - #1211 (@jayeshp19)
fix release - #1176 (@theomonnom)
fix realtime API audio format values - #1092 (@longcw)
Expose usage metrics for Realtime model - #1036 (@yuyuma)
fix: Azure realtime model does not accept null for max_response_output_tokens - #927 (@davidzhao)
fix: Azure realtime model does not accept null for max_response_output_tokens - #927 (@davidzhao)
add update_options to TTS - #922 (@theomonnom)
Groq integration with Whisper-compatible STT endpoints - #986 (@jayeshp19)
pipelineagent: expose timing metrics & api errors wip - #957 (@theomonnom)
openai: fix low timeouts - #926 (@theomonnom)
expose usage metrics - #984 (@theomonnom)
add x.ai support - #907 (@theomonnom)
add x.ai support - #907 (@theomonnom)
Fix functions to include content - #897 (@martin-purplefish)
fix: handle when STT does not return any speech - #854 (@davidzhao)
fix: handle when STT does not return any speech - #854 (@davidzhao)
Support for Realtime API with Azure OpenAI - #848 (@davidzhao)
oai-realtime: fix function calls - #826 (@KillianLucas)
oai-realtime: log response errors - #819 (@theomonnom)
OpenAI Realtime API support - #814 (@theomonnom)
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Nothing published for this version
Your coding agent can read these notes before it upgrades. Set up the MCP server →