List your company

Code Generation on LiveCodeBench v6 (Accuracy, Mean, Drop)

59.06AccuracyBF16
-2.362413.583829.5345.4762May 18, 2026

Evaluation results

MethodLinks
BF16Model=Qwen3-32B, BPE=16.002026.0559.0674.19—
TurboQuant (no MP)Model=GLM-4.7 FP8, 358B, BPE=3.252026.0558.4878.150.26
Saw-INT4Model=Qwen3-32B, BPE=4.252026.0556.0274.430.24
TurboQuant (no MP)Model=Qwen3-32B, BPE=3.252026.0555.5671.99-2.2
OSCARModel=Qwen3-32B, BPE=2.282026.0553.5774.17-0.02
OSCARModel=GLM-4.7 FP8, 358B, BPE=2.282026.0552.6378.160.27
Saw-INT4Model=GLM-4.7 FP8, 358B, BPE=4.252026.0550.4977.950.06
BF16Model=GLM-4.7 FP8, 358B, BPE=16.002026.0549.1277.89—
BF16Model=Qwen3-8B, BPE=16.002026.0549.0170.84—
BF16Model=Qwen3-4B Thinking-2507, BPE=16.002026.0548.6675.64—
QuaRot-INT2Model=GLM-4.7 FP8, 358B, BPE=2.252026.0548.5475.14-2.75
Saw-INT4Model=Qwen3-8B, BPE=4.252026.0547.9569.97-0.87
OSCARModel=Qwen3-8B, BPE=2.282026.0546.3269.416-1.42
Saw-INT4Model=Qwen3-4B Thinking-2507, BPE=4.252026.0546.273.11-2.53
OSCARModel=Qwen3-4B Thinking-2507, BPE=2.282026.0545.3871.864-3.78
Naive INT2Model=GLM-4.7 FP8, 358B, BPE=2.252026.0537.0360.49-17.4
TurboQuant (no MP)Model=Qwen3-8B, BPE=3.252026.0521.0556.88-13.96
TurboQuant (no MP)Model=Qwen3-4B Thinking-2507, BPE=3.252026.050.5831.74-43.9
QuaRot-INT2Model=Qwen3-8B, BPE=2.252026.050.5810.14-60.7
QuaRot-INT2Model=Qwen3-32B, BPE=2.252026.050.397.9-66.29
QuaRot-INT2Model=Qwen3-4B Thinking-2507, BPE=2.252026.0501.4-74.24
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