Set your criteria
Choose what to test, how to score it, and what counts as a pass.
AI evaluation and release gates
Know what passes before you deploy.
Test your agents against your own cases and criteria. Review failures and use the results in your release pipeline.
Open source. Self-hosted.
Choose what to test, how to score it, and what counts as a pass.
Evaluate large datasets, track progress, and inspect or export individual results.
Use the final pass or fail result in CI. Incomplete evaluations keep the release gate closed.