Benchmarks
Reasoning benchmarks
We track 227 benchmarks in this topic. Main benchmarks are listed by how many models they cover, most first. Where a benchmark has separate results for a subset of its tasks, those results appear right under it. Click one for its leaderboard.
Showing 1–100 of 227 benchmarks.
- SnakeBench260 models
- SimpleBench105 models
- Chess Puzzles93 models
- Mystery Game Puzzles85 models
- HELM Lite: LegalBench - EM69 models
- HELM Lite: OpenbookQA - EM69 models
- GeoBench ACW45 models
- LiveBench Reasoning42 models
- Toolathlon40 models
- GeoBench Photos31 models
- BoolQ29 models
- VPCT27 models
- ARC-C25 models
- PIQA24 models
- t2-bench21 models
- ARC-AGI v220 models
- OmniDocBench 1.519 models
- ARC-Easy17 models
- OpenBookQA17 models
- Graphwalks BFS <128k13 models
- BIG-Bench Hard12 models
- Graphwalks BFS >128k12 models
- Graphwalks parents <128k12 models
- COLLIE11 models
- MedXpertQA10 models
- MuirBench10 models
- BIG-Bench Extra Hard9 models
- Bird-SQL (dev)9 models
- EgoSchema9 models
- AA-Briefcase v1.18 models