Theorem Proving on Prover-Bench
70.8Pass@32LongCat-Flash-Prover
Evaluation Results
| Method | Links | |
|---|---|---|
| LongCat-Flash-ProverMode=sketch-proof, Search Strategy=TIR & Tree Search, Budget (b)=2202026.03 | 70.8 | |
| LongCat-Flash-ProverMode=sketch-proof, Search Strategy=TIR, Budget (b)=2202026.03 | 69.5 | |
| LongCat-Flash-ProverEvaluation Mode=sketch-proof mode, TIR (Tool-Integrated Reasoning)=true2026.03 | 66.5 | |
| DeepSeek-Prover-V2-671BBudget (b)=5122026.03 | 59.1 | |
| LongCat-Flash-ProverEvaluation Mode=whole-proof mode, TIR (Tool-Integrated Reasoning)=true2026.03 | 57.9 | |
| Goedel-Prover-V2-32BModel Category=Open-Weights Prover Models, Self-correction=false2026.03 | 53.2 | |
| DeepSeek-Prover-V2-671BModel Category=Open-Weights Prover Models2026.03 | 52.9 | |
| LongCat-Flash-ProverEvaluation Mode=whole-proof mode, TIR (Tool-Integrated Reasoning)=false2026.03 | 49.9 | |
| Goedel-Prover-V2-8BModel Category=Open-Weights Prover Models, Self-correction=false2026.03 | 49.5 | |
| DeepSeek-Prover-V2-7BModel Category=Open-Weights Prover Models2026.03 | 49 | |
| Leanabell-Prover-V2-DSBudget (b)=1282026.03 | 48.7 | |
| Leanabell-Prover-V2-DSModel Category=Open-Weights Prover Models2026.03 | 47.8 | |
| Kimina-Prover-72BModel Category=Open-Weights Prover Models2026.03 | 44.6 | |
| Kimi-K2.5Model Category=Open-Weights Reasoning Models2026.03 | 44.3 | |
| Leanabell-Prover-V2-KMBudget (b)=1282026.03 | 42.9 | |
| DeepSeek-V3.2Model Category=Open-Weights Reasoning Models2026.03 | 42.8 | |
| Leanabell-Prover-V2-KMModel Category=Open-Weights Prover Models2026.03 | 39.8 | |
| Kimina-Prover-8BModel Category=Open-Weights Prover Models2026.03 | 37.8 | |
| Gemini-3 ProModel Category=Close-Weights Reasoning Models2026.03 | 33.5 |