Language Understanding on MMLU (Accuracy, TPS, and Speedup)
86AccuracyGLM-4.7-FP8
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| GLM-4.7-FP8Mode=Standalone2026.05 | 86 | — | 70 | — | |
| PARSETarget Model=GLM-4.7-FP8, Draft Model=Qwen3.5-9B2026.05 | 86 | — | 104.2 | 1.49 | |
| Qwen3.5-9BMode=Standalone2026.05 | 78 | — | — | — |