v0.13.1

Key Changes

Any-LLM extension

Since this version, the extension module includes a new any-llm adapter as well. Please refer to the updated documentation and examples at https://github.com/openai/openai-agents-python/tree/main/examples/model_providers

What's Changed

• feat: add any-llm model support with responses-compatible routing by @seratch in #2706
• fix: preserve static MCP meta in converted function tools by @seratch in #2769
• fix: #2760 wait for realtime response.done before follow-up response.create by @seratch in #2763
• fix: handle cancelled single function tools as tool failures by @elainegan-openai in #2762
• fix: optionize initialized notification tolerance by @elainegan-openai in #2765
• fix: remove duplicate CompactionItem from RunItem union by @KanchiShimono in #2761

Documentation & Other Changes

• docs: add 0.13 changelog by @seratch in #2744
• docs: update translated document pages by @github-actions[bot] in #2759
• fix: harden example auto-runs against PATH and port conflicts by @seratch in #2770
• Release 0.13.1 by @github-actions[bot] in #2768

Full Changelog: v0.13.0...v0.13.1