Every model's score on every benchmark we track, each linked to where it was published. Independent runs and lab-reported numbers are kept apart. Many of these benchmarks feed the Respan Index.
177 of 177 benchmarks
Math
21 models, higher is better
Problems from the American Invitational Mathematics Examination, a competition-level math benchmark used to test multi-step quantitative reasoning.
Benchmark data from llmmetric.com, Respan's model data service.