pydantic-ai
https://github.com/pydantic/pydantic-ai
Python
GenAI Agent Framework, the Pydantic way
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
Python not yet supported4 Subscribers
Add a CodeTriage badge to pydantic-ai
Help out
- Issues
- Dataset.evaluate never extracts span-tree metrics when the task raises
- Let a capability declare that it must be the innermost execution wrapper, and have the sorter enforce it
- Accept non-string `metadata` values from OpenAI-compatible chat completions
- CI Review agent job fails to check out the PR head: pre-warm rewrites pydantic-ai-runner.lock
- Bump the python-packages group across 1 directory with 16 updates
- Events from tool calls that trail a streamed final output never reach `event_stream_handler`
- tests: pin Bedrock thinking + auto-output silent downgrade (native/prompted) path
- [roundtrip-sweep] ModelMessagesTypeAdapter: LoadCapabilityReturnPart.content silently drops unknown keys on round-trip (gap in PR #7933 fix)
- `TestModel` structured output retries out on `datetime`, `UUID`, `time`, and URL fields
- `OpenAIChatModel` rejects Nebius responses because `metadata.weight_versions` is a list
- Docs
- Python not yet supported