List your company
ResearchFormerly WizWand

Mathematical Reasoning on GSM-Infinite Hard

50.4AccuracyDeepSeek-V3.2
7.13618.36829.640.832Jan 6, 2026
Updated 8mo ago

Evaluation results

MethodLinks
DeepSeek-V3.2Length=16K, #Activated Params=37B, #Total Params=671B2026.0150.4
DeepSeek-V3.2Length=32K, #Activated Params=37B, #Total Params=671B2026.0145.2
DeepSeek-V3.1Length=16K, #Activated Params=37B, #Total Params=671B2026.0141.5
DeepSeek-V3.1Length=32K, #Activated Params=37B, #Total Params=671B2026.0138.8
MiMo-V2-FlashLength=16K, #Activated Params=15B, #Total Params=309B2026.0137.7
DeepSeek-V3.1Length=64K, #Activated Params=37B, #Total Params=671B2026.0134.7
Kimi-K2Length=16K, #Activated Params=32B, #Total Params=1043B2026.0134.6
MiMo-V2-FlashLength=32K, #Activated Params=15B, #Total Params=309B2026.0133.7
DeepSeek-V3.2Length=64K, #Activated Params=37B, #Total Params=671B2026.0132.6
MiMo-V2-FlashLength=64K, #Activated Params=15B, #Total Params=309B2026.0131.5
MiMo-V2-FlashLength=128K, #Activated Params=15B, #Total Params=309B2026.0129
DeepSeek-V3.1Length=128K, #Activated Params=37B, #Total Params=671B2026.0128.7
Kimi-K2Length=32K, #Activated Params=32B, #Total Params=1043B2026.0126.1
DeepSeek-V3.2Length=128K, #Activated Params=37B, #Total Params=671B2026.0125.7
Kimi-K2Length=64K, #Activated Params=32B, #Total Params=1043B2026.0116
Kimi-K2Length=128K, #Activated Params=32B, #Total Params=1043B2026.018.8