How do evals work?
Scorecards, judges, datasets, benchmarks, traces, and dashboards in one suite. Every run can be scored, and a read-only API exposes the results.
The Evals suite holds the pieces of a quality program: scorecards (your rubrics), judges that apply them, datasets and golden sets, benchmarks that compare configurations, the traces of every run, and dashboards that show pass rates over time. Feedback from reviewers feeds back into the same records.
Scores are attached to the run they judge, so a reviewer can open the trace and see why. A read-only REST API exposes traces, scores, and analytics for your own reporting. See Evals.
Written for developers, it and security. Last reviewed 2026-09-15.
Related questions
Still need an answer?
Tell us what you were looking for and we reply within one business day.
Contact the team →Running a security review?
Documents, controls, and the request form live in the Trust Center.
Open the Trust Center →