Chinese-language ability on CMATH
84.8AccuracyGLM-4.6V-FlashX
Evaluation Results
| Method | Links | |
|---|---|---|
| GLM-4.6V-FlashX2026.01 | 84.8 | |
| Ovis2.5Parameter Scale=9B2026.01 | 83.4 | |
| Qwen2.5-VLParameter Scale=72B2026.01 | 74.8 | |
| Qwen3-VLParameter Scale=8B2026.01 | 66.2 | |
| Ostrakon-VLParameter Scale=8B2026.01 | 63.5 | |
| InternVL3.5Parameter Scale=8B2026.01 | 62 |