List your company

Mathematical Reasoning on GSM8K (p@1 and p@16)

96.66p@1 AccuracyST
94.0694.73595.4196.085Feb 2, 2026

Evaluation results

MethodLinks
STModel=Qwen3-30B-A3B-Thinking2026.0296.6696.89
CoTModel=Qwen3-30B-A3B-Thinking2026.0296.597.88
DoLaModel=Qwen3-30B-A3B-Thinking2026.0296.597.88
ST-GModel=Qwen3-30B-A3B-Thinking2026.0296.597.88
LEDModel=Qwen3-30B-A3B-Thinking2026.0296.4897.73
STModel=Qwen3-4B-Thinking2026.0295.2596.59
CoTModel=Qwen3-4B-Thinking2026.0295.297.35
ST-GModel=Qwen3-4B-Thinking2026.0295.297.19
LEDModel=Qwen3-4B-Thinking2026.0295.297.35
DoLaModel=Qwen3-4B-Thinking2026.0295.1197.35
STModel=MiMo-7B-RL2026.0294.796.36
LEDModel=MiMo-7B-RL2026.0294.3696.89
ST-GModel=MiMo-7B-RL2026.0294.3397.04
DoLaModel=MiMo-7B-RL2026.0294.2497.04
CoTModel=MiMo-7B-RL2026.0294.1697.19