List your company

Mathematical Reasoning on AIME25 (Score)

66.67ScoreMCTS-RAG
18.133230.734143.33555.9359Apr 7, 2026

Evaluation results

MethodLinks
MCTS-RAGModel=Qwen3-30B-A3B-Instruct-25072026.0466.67
PRISM-MCTSModel=Qwen3-30B-A3B-Instruct-25072026.0463.33
PRISM-MCTSModel=GPT-4.1-mini2026.0453.33
Zero-shotModel=Qwen3-30B-A3B-Instruct-25072026.0453.33
ReActModel=GPT-4.1-mini2026.0450
MCTS-RAGModel=GPT-4.1-mini2026.0450
ReActModel=Qwen3-30B-A3B-Instruct-25072026.0450
Search-O1Model=Qwen3-30B-A3B-Instruct-25072026.0450
Zero-shotModel=GPT-4.1-mini2026.0448.67
Search-O1Model=GPT-4.1-mini2026.0443.33
ReST-MCTS*Model=GPT-4.1-mini2026.0420
ReST-MCTS*Model=Qwen3-30B-A3B-Instruct-25072026.0420