List your company

General Evaluation on AGIEval

70.22AccuracyDeepSeek-R1 (BF16)
23.6835.762547.84559.9275Nov 28, 2025Dec 14, 2025Dec 31, 2025Jan 17, 2026Feb 3, 2026Feb 20, 2026Mar 9, 2026

Evaluation results

MethodLinks
DeepSeek-R1 (BF16)Model=DeepSeek-R1, Setting=W16A16, Sparsity=0, Evaluation Protocol=Generative2025.1270.22
SQ-formatModel=DeepSeek-R1, Setting=W(SQ5)A8, Sparsity=0.875, Evaluation Protocol=Generative2025.1269.58
FineRMoEBase Model=Qwen2.5-7B, #P/B=26.65, #AP/B=7.942026.0356.7
PTBase Model=Qwen2.5-7B, #P/B=7.62, #AP/B=7.622026.0356.2
C32A2Base Model=Qwen2.5-7B, #P/B=184.42, #AP/B=13.332026.0355.91
CTBase Model=Qwen2.5-7B, #P/B=7.62, #AP/B=7.622026.0352.67
NVShardBase Model=Qwen2.5-7B, #P/B=47.55, #AP/B=7.632026.0349.36
HSA-ULTraining Strategy=Annealing, Architecture=MoE, Total Params=8B, Activated Params=1B, Training Tokens=8T2025.1144.08
C32A2Base Model=Qwen2.5-1.5B, #P/B=37.62, #AP/B=2.942026.0343.26
FineRMoEBase Model=Qwen2.5-1.5B, #P/B=5.64, #AP/B=1.852026.0343.14
PTBase Model=Qwen2.5-1.5B, #P/B=1.54, #AP/B=1.542026.0339.83
CTBase Model=Qwen2.5-1.5B, #P/B=1.54, #AP/B=1.542026.0338.91
TRM-MoETraining Strategy=Base, Architecture=MoE, Total Params=8B, Activated Params=1B, Training Tokens=8T2025.1138.64
NVShardBase Model=Qwen2.5-1.5B, #P/B=9.88, #AP/B=1.782026.0336.78
HSA-ULTraining Strategy=Base, Architecture=MoE, Total Params=8B, Activated Params=1B, Training Tokens=8T2025.1136.52
Qwen3Training Strategy=Annealing, Architecture=Dense, Total Params=0.6B, Activated Params=0.6B, Training Tokens=36T2025.1133.58
Qwen2.5Training Strategy=Annealing, Architecture=Dense, Total Params=0.5B, Activated Params=0.5B, Training Tokens=18T2025.1130.01
HSA-ULTraining Strategy=Annealing, Architecture=Dense, Total Params=0.5B, Activated Params=0.5B, Training Tokens=4T2025.1129.29
S16A4Base Model=Qwen2.5-7B, #P/B=7.62, #AP/B=3.342026.0328.94
C32A2Base Model=Qwen2.5-0.5B, #P/B=10.36, #AP/B=0.952026.0328.31
FineRMoEBase Model=Qwen2.5-0.5B, #P/B=1.68, #AP/B=0.652026.0328.19
PTBase Model=Qwen2.5-0.5B, #P/B=0.49, #AP/B=0.492026.0328.07
CTBase Model=Qwen2.5-0.5B, #P/B=0.49, #AP/B=0.492026.0327.31
NVShardBase Model=Qwen2.5-0.5B, #P/B=2.83, #AP/B=0.632026.0326.79
S16A4Base Model=Qwen2.5-1.5B, #P/B=1.78, #AP/B=0.912026.0325.91
DUBase Model=Qwen2.5-0.5B, #P/B=2.83, #AP/B=0.942026.0325.86
DUBase Model=Qwen2.5-7B, #P/B=47.54, #AP/B=13.322026.0325.65
DUBase Model=Qwen2.5-1.5B, #P/B=9.87, #AP/B=2.932026.0325.52
S16A4Base Model=Qwen2.5-0.5B, #P/B=0.63, #AP/B=0.402026.0325.47