v2.3.21

agno-agi/agnov2.3.21Dec 23, 2025by manuhortet

AI Summary

Added full support for AgentAsJudge evaluations on AgentOS and fixed RunInput parsing issues with Message objects.

Key Highlights

AgentAsJudge Evals are now fully supported on AgentOS
Fixed RunInput object persistence when provided as a list of Message objects

New Features

AgentAsJudge Evals Support

Full Release Notes

# Changelog

## New Features:

- **Agent as Judge on AgentOS**: Our new [[AgentAsJudge Evals](https://docs.agno.com/basics/evals/agent-as-judge/overview)](https://docs.agno.com/basics/evals/agent-as-judge/overview) are now completely supported on the AgentOS: you can configure and trigger new runs, and will see existing runs listed with the rest in the Evals page.

## Bug Fixes:

- **RunInput parsing**: Fixed an issue persisting RunInput objects when providing it as a list of Message objects.
- **Mistral Embedder**: Fixed an issue setting timeout values for the MistralEmbedder class.

## What's Changed
* cookbook: Speech to text workflow example by @willemcdejongh in https://github.com/agno-agi/agno/pull/5742
* feat: return agent-as-judge evals on GET endpoints [SDK-403] by @manuhortet in https://github.com/agno-agi/agno/pull/5663
* [fix] Make RunInput input_content JSON-serializable by @liqiongyu in https://github.com/agno-agi/agno/pull/5810
* chore: v2.3.21 by @manuhortet in https://github.com/agno-agi/agno/pull/5840


**Full Changelog**: https://github.com/agno-agi/agno/compare/v2.3.20...v2.3.21