List your company

Mathematical Reasoning on AMO-Bench (accuracy)

59.8AMO-Bench AccuracySU-01
38.4844.01549.5555.085May 13, 2026

Evaluation results

MethodLinks
SU-01Number of runs=82026.0559.8
Qwen3.6-35B-A3BNumber of runs=82026.0558.8
GLM-4.7-FlashNumber of runs=82026.0553.8
P1-30B-A3BNumber of runs=82026.0541.3
Nemotron-Cascade-2Number of runs=82026.0540.8
Gemma-4-31BNumber of runs=82026.0539.3