List your company

Math Problem Solving on LiveMathBench

100AIME 24 ScoreFoT (Round 2)
4861.57588.5Apr 18, 2026

Evaluation results

MethodLinks
FoT (Round 2)Base Model=Gemini-3-Pro-Preview2026.0410010093.590.910090.995.27292.8
Isolated AgentBase Model=Gemini-3-Pro-Preview2026.0496.793.393.586.488.972.776.26984.6
Isolated AgentBase Model=DeepSeek-R1-Distill-Qwen-7B2026.04504067.484.172.227.352.43653.7
FoT (Round 3)Base Model=DeepSeek-R1-Distill-Qwen-7B2026.04504071.784.172.236.452.43655.3