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
Reasoning
34 models, higher is better
A very hard, broad exam of expert-level questions across many fields, designed to remain difficult for frontier models.
Benchmark data from llmmetric.com, Respan's model data service.