List your company

Scientific Reasoning on GPQA (Accuracy, Mean, Drop)

73.74AccuracySaw-INT4
-2.949616.960236.8756.7798May 18, 2026

Evaluation results

MethodLinks
Saw-INT4Model=GLM-4.7 FP8, 358B, BPE=4.252026.0573.7477.950.06
OSCARModel=GLM-4.7 FP8, 358B, BPE=2.282026.0573.5778.160.27
BF16Model=GLM-4.7 FP8, 358B, BPE=16.002026.0573.2377.89—
QuaRot-INT2Model=GLM-4.7 FP8, 358B, BPE=2.252026.0568.0175.14-2.75
BF16Model=Qwen3-4B Thinking-2507, BPE=16.002026.0567.2775.64—
TurboQuant (no MP)Model=GLM-4.7 FP8, 358B, BPE=3.252026.0566.6778.150.26
Saw-INT4Model=Qwen3-4B Thinking-2507, BPE=4.252026.0566.3773.11-2.53
OSCARModel=Qwen3-4B Thinking-2507, BPE=2.282026.0564.9571.864-3.78
OSCARModel=Qwen3-32B, BPE=2.282026.0560.474.17-0.02
Saw-INT4Model=Qwen3-32B, BPE=4.252026.0559.2974.430.24
TurboQuant (no MP)Model=Qwen3-32B, BPE=3.252026.0558.6971.99-2.2
BF16Model=Qwen3-32B, BPE=16.002026.0558.4974.19—
BF16Model=Qwen3-8B, BPE=16.002026.0556.6770.84—
TurboQuant (no MP)Model=Qwen3-8B, BPE=3.252026.0555.0556.88-13.96
OSCARModel=Qwen3-8B, BPE=2.282026.0555.0569.416-1.42
Saw-INT4Model=Qwen3-8B, BPE=4.252026.0554.8569.97-0.87
Naive INT2Model=GLM-4.7 FP8, 358B, BPE=2.252026.0554.5560.49-17.4
TurboQuant (no MP)Model=Qwen3-4B Thinking-2507, BPE=3.252026.0541.4131.74-43.9
QuaRot-INT2Model=Qwen3-32B, BPE=2.252026.0519.77.9-66.29
QuaRot-INT2Model=Qwen3-8B, BPE=2.252026.0514.9810.14-60.7
QuaRot-INT2Model=Qwen3-4B Thinking-2507, BPE=2.252026.050.341.4-74.24
Naive INT2Model=Qwen3-4B Thinking-2507, BPE=2.252026.0500-75.64
Naive INT2Model=Qwen3-8B, BPE=2.252026.0500-70.84
Naive INT2Model=Qwen3-32B, BPE=2.252026.0500-74.19