Starstruk Logo MarkStarstruk

v1.58.0

pydantic/pydantic-aiv1.58.0Feb 11, 2026by DouweM

AI Summary

Enhances evaluation capabilities with report-level evaluators and experiment-wide analyses, and adds extra headers for Google provider.

Key Highlights

  • Report-level evaluators & experiment-wide analyses
  • Multi-run aggregation support (repeat parameter)
  • Added `extra_headers` support for Google model provider

New Features

  • Report-level evaluators
  • Experiment-wide analyses
  • Multi-run aggregation
  • Google extra headers support

Full Release Notes

<!-- Release notes generated using configuration in .github/release.yml at main -->

## What's Changed
### 🚀 Features
* Report-level evaluators & experiment-wide analyses by @dmontagu in https://github.com/pydantic/pydantic-ai/pull/4243
* Add multi-run aggregation support (repeat parameter) to pydantic-evals by @dmontagu in https://github.com/pydantic/pydantic-ai/pull/4253
* Add `extra_headers` support for Google model provider by @saakshigupta2002 in https://github.com/pydantic/pydantic-ai/pull/4241

## New Contributors
* @saakshigupta2002 made their first contribution in https://github.com/pydantic/pydantic-ai/pull/4241

**Full Changelog**: https://github.com/pydantic/pydantic-ai/compare/v1.57.0...v1.58.0