List your company

Mathematical Problem Solving on AMC (Accuracy %)

96.39AMC AccuracyMajority Vote
91.065292.447693.8395.2124Apr 18, 2026

Evaluation results

MethodLinks
Majority VoteModel Pool=Gemini-2.5-flash2026.0496.39
Majority VoteModel Pool=Qwen3-235B-A22B-Instruct-25072026.0495.18
Majority VoteModel Pool=GPT-OSS-120B2026.0495.18
BaseModel Pool=Qwen3-235B-A22B-Instruct-25072026.0494.73
Self RefineModel Pool=Qwen3-235B-A22B-Instruct-25072026.0494.58
ParaManagerStrategy=Model Orchestration2026.0494.43
Self RefineModel Pool=Gemini-2.5-flash2026.0493.98
Self RefineModel Pool=GPT-OSS-120B2026.0492.77
ToolOrchestraStrategy=Model Orchestration2026.0492.62
BaseModel Pool=Gemini-2.5-flash2026.0492.47
PuppeteerStrategy=Model Orchestration2026.0491.87
BaseModel Pool=GPT-OSS-120B2026.0491.27