List your company

Mathematical Reasoning on AIME 2025 (p@1, p@16)

82.08P@1DoLa
53.698461.066768.43575.8033Feb 2, 2026

Evaluation results

MethodLinks
DoLaModel=Qwen3-30B-A3B-Thinking2026.0282.0893.33
CoTModel=Qwen3-30B-A3B-Thinking2026.0281.2590
LEDModel=Qwen3-30B-A3B-Thinking2026.0281.0493.33
ST-GModel=Qwen3-30B-A3B-Thinking2026.0279.7993.33
LEDModel=Qwen3-4B-Thinking2026.0276.4690
STModel=Qwen3-30B-A3B-Thinking2026.0275.2190
CoTModel=Qwen3-4B-Thinking2026.0274.1790
STModel=Qwen3-4B-Thinking2026.0272.2983.33
DoLaModel=Qwen3-4B-Thinking2026.0272.0886.67
ST-GModel=Qwen3-4B-Thinking2026.0267.590
LEDModel=MiMo-7B-RL2026.0260.6283.33
CoTModel=MiMo-7B-RL2026.0259.1780
DoLaModel=MiMo-7B-RL2026.0259.1780
ST-GModel=MiMo-7B-RL2026.0259.1780
STModel=MiMo-7B-RL2026.0254.7980