List your company

Mathematical Reasoning on AIME 25 (Pass@1, #Tok)

80Pass@1 AccuracyCoD
7.226.14563.9Jun 5, 2026

Evaluation results

MethodLinks
CoDBackbone=Qwen3-4B-Thinking-25072026.068021,246
BaselineBackbone=Qwen3-4B-Thinking-25072026.0676.722,708
DyConBackbone=Qwen3-4B-Thinking-25072026.0676.721,100
NothinkingBackbone=Qwen3-4B-Thinking-25072026.0673.319,177
DEERBackbone=Qwen3-4B-Thinking-25072026.067013,342
NothinkingBackbone=QwQ-32B2026.067015,312
BaselineBackbone=Qwen3-14B2026.067016,613
DyConBackbone=Qwen3-14B2026.067014,537
DyConBackbone=QwQ-32B2026.0666.713,640
DEERBackbone=Qwen3-14B2026.0666.711,135
BaselineBackbone=QwQ-32B2026.066016,462
Dynasor-CoTBackbone=Qwen3-14B2026.066012,159
NoWaitBackbone=Qwen3-14B2026.066012,276
SEALBackbone=QwQ-32B2026.0656.712,089
NoWaitBackbone=QwQ-32B2026.0656.711,871
NoWaitBackbone=Qwen3-4B-Thinking-25072026.0653.313,322
DEERBackbone=QwQ-32B2026.065011,598
CoDBackbone=QwQ-32B2026.0646.713,289
CoDBackbone=Qwen3-14B2026.0646.712,391
TrimRBackbone=QwQ-32B2026.0643.38,827
Controlling Thinking SpeedBackbone=DeepSeek-R1-Distill-Qwen-7B2026.064010,997
FlashThinkBackbone=QwQ-32B2026.064011,861
BaselineBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0636.715,245
DEERBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0636.77,257
DyConBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0636.712,415
CoDBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0633.314,333
ThinkpilotBackbone=Qwen3-4B-Thinking-25072026.06308,814
SEALBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0626.711,092
NoWaitBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0626.79,302
NothinkingBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0623.34,385
NothinkingBackbone=Qwen3-14B2026.0623.35,115
ThinkpilotBackbone=Qwen3-14B2026.0623.33,488
ThinkpilotBackbone=DeepSeek-R1-Distill-Qwen-7B2026.06101,961