List your company

Mathematical Reasoning on MATH 500 (Accuracy and Token Usage)

97Accuracy (%)COT
66.63274.51682.490.284Apr 9, 2026

Evaluation results

MethodLinks
COTModel=QwQ-32B2026.04974,025
SATModel=QwQ-32B2026.04973,256
SATModel=Qwen3-14B2026.0496.62,904
SATModel=Qwen3-32B2026.0496.62,719
COTModel=Qwen3-14B2026.0496.24,598
SATModel=Qwen3-8B2026.0495.83,215
SATModel=Qwen3-4B2026.0495.62,833
COTModel=Qwen3-8B2026.0495.65,166
COTModel=Qwen3-32B2026.0495.64,358
COTModel=Qwen3-4B2026.04954,823
SATModel=DS-Qwen-14B2026.0494.82,515
DEERModel=QwQ-32B2026.0494.63,316
CGRSModel=Qwen3-14B2026.0494.53,235
COTModel=DS-Qwen-14B2026.0494.43,539
DEERModel=Qwen3-32B2026.0494.23,418
NoThinkingModel=QwQ-32B2026.0494.24,276
CGRSModel=QwQ-32B2026.0494.22,810
DEERModel=Qwen3-14B2026.04943,074
SATModel=Nemo-Llama-8b2026.04942,844
TALEModel=QwQ-32B2026.04943,533
DynasorModel=QwQ-32B2026.04944,156
COTModel=Nemo-Llama-8b2026.0493.83,412
TALEModel=Qwen3-14B2026.0493.73,389
TALEModel=Qwen3-32B2026.0493.63,857
CGRSModel=Qwen3-8B2026.0493.33,507
CGRSModel=Qwen3-32B2026.0493.12,993
COTModel=DS-Qwen-7B2026.0492.83,537
SATModel=DS-Qwen-7B2026.0492.82,237
ThinkSwitcherModel=DS-Qwen-14B2026.0492.73,572
DEERModel=Qwen3-4B2026.0492.63,214
DEERModel=Qwen3-8B2026.0492.62,732
TALEModel=Qwen3-8B2026.0492.33,885
DynasorModel=Qwen3-8B2026.0491.73,841
DEERModel=Nemo-Llama-8b2026.0491.42,995
CGRSModel=Qwen3-4B2026.0491.32,704
ThinkSwitcherModel=DS-Qwen-7B2026.0491.33,495
DynasorModel=Qwen3-4B2026.0490.13,877
DEERModel=DS-Qwen-7B2026.0489.82,143
DEERModel=DS-Qwen-14B2026.0489.82,577
TALEModel=Qwen3-4B2026.0489.12,657
TALEModel=DS-Qwen-7B2026.0489.12,657
CGRSModel=DS-Qwen-7B2026.0487.61,867
NoThinkingModel=Qwen3-8B2026.0487.11,239
NoThinkingModel=Qwen3-14B2026.0487853
NoThinkingModel=Qwen3-32B2026.04871,054
DynasorModel=Qwen3-32B2026.0485.23,486
NoThinkingModel=Qwen3-4B2026.0484.9988
DynasorModel=Qwen3-14B2026.0484.43,667
ThinkSwitcherModel=DS-Qwen-1.5B2026.0482.44,544
SATModel=DS-Qwen-1.5B2026.0482.43,230
COTModel=DS-Qwen-1.5B2026.0482.24,723
DynasorModel=DS-Qwen-7B2026.0481.82,070
NoThinkingModel=DS-Qwen-7B2026.0480.91,173
DEERModel=DS-Qwen-1.5B2026.0467.82,497