Latest finalised run per model
Legal AI Leaderboard
Ranked by total rubric score. Domain, difficulty, trap performance, cost, and the original auditable JSON remain visible.
Top score0.0%
Models ranked0
Average score0.0%
Average trap accuracy0%
Independent leaderboard: results shown here were published by this server’s administrator. Anyone may run the benchmark and download an auditable JSON copy.
Model ranking
Only each model’s most recently finalised run is ranked. Ties favour lower estimated cost.
No finalised scores yetRun the 100-case benchmark, review all grades, and finalise it while logged in as this server’s administrator.
Run the benchmark
Run the benchmark