List your company

Mathematical Reasoning on GSM8K (test) (AVG@8, PASS@8)

96.36AVG@8IAC
93.583294.304195.02595.7459Jan 22, 2026

Evaluation results

MethodLinks
IACModel=Qwen3-14B2026.0196.3698.11
RLOOModel=Qwen3-14B2026.0196.0497.43
GRPOModel=Qwen3-14B2026.0195.9697.53
DrGRPOModel=Qwen3-14B2026.0195.9697.53
Reinforce++Model=Qwen3-14B2026.0195.8297.79
IACModel=Qwen3-8B2026.0195.7297.56
GRPOModel=Qwen3-8B2026.0195.5996.51
RLOOModel=Qwen3-8B2026.0195.5697.25
Reinforce++Model=Qwen3-8B2026.0195.5696.96
DrGRPOModel=Qwen3-8B2026.0195.3597.11
IACModel=Qwen3-4B2026.0194.6397.38
DrGRPOModel=Qwen3-4B2026.0194.196.43
Reinforce++Model=Qwen3-4B2026.0194.0496.45
GRPOModel=Qwen3-4B2026.0193.7296.74
RLOOModel=Qwen3-4B2026.0193.6996.98