List your company

Mathematical Problem Solving on HMMT 2025

76.67AccuracyMajority Vote
28.341240.888153.43565.9819Apr 18, 2026Apr 26, 2026May 4, 2026May 13, 2026May 21, 2026May 29, 2026Jun 7, 2026

Evaluation results

MethodLinks
Majority VoteModel Pool=Gemini-2.5-flash2026.0476.67
ParaManagerStrategy=Model Orchestration2026.0475
Self RefineModel Pool=Gemini-2.5-flash2026.0473.33
PuppeteerStrategy=Model Orchestration2026.0473.33
ToolOrchestraStrategy=Model Orchestration2026.0472.92
Majority VoteModel Pool=GPT-OSS-120B2026.0466.67
BaseModel Pool=Gemini-2.5-flash2026.0460.83
Self RefineModel Pool=GPT-OSS-120B2026.0460
pass@32k=322026.0656.7
Self RefineModel Pool=Qwen3-235B-A22B-Instruct-25072026.0450.83
BaseModel Pool=Qwen3-235B-A22B-Instruct-25072026.0450.42
Majority VoteModel Pool=Qwen3-235B-A22B-Instruct-25072026.0450
BaseModel Pool=GPT-OSS-120B2026.0448.33
iS@1k=12026.0635
SCsamples=322026.0630.8
pass@1k=12026.0630.2