Mathematical Reasoning on AIME25 (Score)
66.67ScoreMCTS-RAG
Evaluation Results
| Method | Links | |
|---|---|---|
| MCTS-RAGModel=Qwen3-30B-A3B-Instruct-25072026.04 | 66.67 | |
| PRISM-MCTSModel=Qwen3-30B-A3B-Instruct-25072026.04 | 63.33 | |
| PRISM-MCTSModel=GPT-4.1-mini2026.04 | 53.33 | |
| Zero-shotModel=Qwen3-30B-A3B-Instruct-25072026.04 | 53.33 | |
| ReActModel=GPT-4.1-mini2026.04 | 50 | |
| MCTS-RAGModel=GPT-4.1-mini2026.04 | 50 | |
| ReActModel=Qwen3-30B-A3B-Instruct-25072026.04 | 50 | |
| Search-O1Model=Qwen3-30B-A3B-Instruct-25072026.04 | 50 | |
| Zero-shotModel=GPT-4.1-mini2026.04 | 48.67 | |
| Search-O1Model=GPT-4.1-mini2026.04 | 43.33 | |
| ReST-MCTS*Model=GPT-4.1-mini2026.04 | 20 | |
| ReST-MCTS*Model=Qwen3-30B-A3B-Instruct-25072026.04 | 20 |