List your company

Math Reasoning on AIME 25 (Pass@1, #Tokens)

66.7Pass@1NoThinking
11.16425.5824054.418Mar 12, 2026

Evaluation results

MethodLinks
NoThinkingBackbone=QwQ-32B2026.0366.711,839
NoWaitBackbone=QwQ-32B2026.0363.38,970
ReBalanceBackbone=QwQ-32B2026.0363.311,575
Dynasor-CoTBackbone=Qwen3-14B2026.036012,159
BaselineBackbone=Qwen3-14B2026.0356.713,125
DEERBackbone=Qwen3-14B2026.0356.711,806
NoWaitBackbone=Qwen3-14B2026.0356.710,924
ReBalanceBackbone=Qwen3-14B2026.0356.711,057
SEALBackbone=QwQ-32B2026.0356.711,384
CoDBackbone=Qwen3-14B2026.0353.311,828
NoThinkingBackbone=Qwen3-14B2026.0353.39,892
BaselineBackbone=QwQ-32B2026.0353.312,627
SEALBackbone=Qwen3-14B2026.035010,901
CoDBackbone=QwQ-32B2026.0346.712,189
DEERBackbone=QwQ-32B2026.0346.710,972
TrimRBackbone=QwQ-32B2026.0343.38,827
DEERBackbone=DeepSeek-R1-Distill-Qwen-7B2026.03408,919
ReBalanceBackbone=DeepSeek-R1-Distill-Qwen-7B2026.034011,074
FlashThinkBackbone=QwQ-32B2026.034011,861
CoDBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0336.710,198
Dynasor-CoTBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0333.311,069
BaselineBackbone=DeepSeek-R1-Distill-Qwen-7B2026.033011,923
Dynasor-CoTBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0326.712,462
SEALBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0326.78,544
ReBalanceBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0326.710,672
NoWaitBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0326.76,979
SEALBackbone=DeepSeek-R1-Distill-Qwen-7B2026.0326.79,835
DEERBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0323.38,719
BaselineBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.032012,739
NoThinkingBackbone=DeepSeek-R1-Distill-Qwen-7B2026.03207,850
NoThinkingBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0316.67,473
NoWaitBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0316.67,574
CoDBackbone=DeepSeek-R1-Distill-Qwen-1.5B2026.0313.310,941