List your company

Mathematical Reasoning on AIME24 (Pass@1, Pass@8, #Invoc.)

73.47Pass@1DART
15.115630.265345.41560.5647Jan 13, 2026

Evaluation results

MethodLinks
DARTBackbone=Qwen3-4B-Thinking-2507, Paradigm=RL, CoT Type=tool-integrated long2026.0173.4791.3710.68
ToRL-ThinkBackbone=Qwen3-4B-Thinking-2507, Paradigm=RL, CoT Type=pure-text long2026.0171.2582.140
UTIRBackbone=Qwen3-4B-Thinking-2507, Paradigm=RL, CoT Type=tool-integrated long2026.0170.5687.6110.08
DAPOBackbone=Qwen3-4B-Thinking-2507, Paradigm=RL, CoT Type=pure-text long2026.0169.1785.670
UTIRBackbone=Qwen3-8B, Paradigm=RL, CoT Type=tool-integrated long2026.0169.1284.5911.97
DARTBackbone=Qwen3-8B, Paradigm=RL, CoT Type=tool-integrated long2026.0168.8986.8112.82
DAPOBackbone=Qwen3-8B, Paradigm=RL, CoT Type=pure-text long2026.0166.1181.440
ToRL-ThinkBackbone=Qwen3-8B, Paradigm=RL, CoT Type=pure-text long2026.0165.782.680
Original checkpointBackbone=Qwen3-8B, CoT Type=pure-text long2026.0164.1779.280
Original checkpointBackbone=Qwen3-4B-Thinking-2507, CoT Type=pure-text long2026.0152.2271.930
STARTBackbone=Qwen3-4B-Thinking-2507, Paradigm=SFT, CoT Type=tool-integrated long2026.0144.3165.1513.3
ToRL-BaseBackbone=Qwen3-8B, Paradigm=RL, CoT Type=tool-integrated short2026.0142.6454.953.86
ToRL-BaseBackbone=Qwen3-4B-Thinking-2507, Paradigm=RL, CoT Type=tool-integrated short2026.0140.4950.733.12
STARTBackbone=Qwen3-8B, Paradigm=SFT, CoT Type=tool-integrated long2026.0135.8358.2215.08
RetoolBackbone=Qwen3-8B, Paradigm=SFT+RL, CoT Type=tool-integrated short2026.0125.4146.731.67
RetoolBackbone=Qwen3-4B-Thinking-2507, Paradigm=SFT+RL, CoT Type=tool-integrated short2026.0123.3335.461.85
Retool (only SFT)Backbone=Qwen3-8B, Paradigm=SFT, CoT Type=tool-integrated short2026.0122.7843.462
Retool (only SFT)Backbone=Qwen3-4B-Thinking-2507, Paradigm=SFT, CoT Type=tool-integrated short2026.0117.3634.142.12