List your company

Mathematical Reasoning on AIME 2024 (Pass@1, Token Usage)

67.33Pass@1 AccuracyMixReasoning
25.7336.5347.3358.13Oct 7, 2025

Evaluation results

MethodLinks
MixReasoningType=ADAPTIVE2025.1067.338,873
BeConciseType=SHORT COT2025.1066.6711,371
DEERType=SHORT COT2025.1066.3310,298
O1-PrunerType=SHORT COT2025.1065.6710,498
NoWaitType=SHORT COT2025.1064.839,936
Qwen3-8BType=BASE LLM2025.1064.6711,975
CoT-ValveType=SHORT COT2025.1033.337,412
ThinklessType=HYBRID2025.1027.337,099