Text Chinese Language Understanding on C3
98.7ScoreGPT-5.4
Evaluation results
| Method | Links | |
|---|---|---|
| GPT-5.4Parameter Count=≈1.5T2026.06 | 98.7 | |
| Opus-4.6Parameter Count=≈800B2026.06 | 98.3 | |
| Qwen3.5-PlusParameter Count=397B2026.06 | 97.9 | |
| K2.5Parameter Count=≈1T2026.06 | 97.6 | |
| Qwen3-VL-32BDecoding Mode=Instruct, Model Scale=32B2026.06 | 96.7 | |
| Qwen3-VL-32BDecoding Mode=Thinking, Model Scale=32B2026.06 | 96.6 | |
| Yuvion VL-32BDecoding Mode=Reasoning, Model Scale=32B2026.06 | 96.2 | |
| Yuvion VL-32BDecoding Mode=Instruct, Model Scale=32B2026.06 | 95.9 | |
| Qwen3-VL-8BDecoding Mode=Thinking, Model Scale=8B2026.06 | 94.9 | |
| Qwen3-VL-8BDecoding Mode=Instruct, Model Scale=8B2026.06 | 93.7 | |
| Yuvion VL-8BDecoding Mode=Instruct, Model Scale=8B2026.06 | 92 | |
| Yuvion VL-8BDecoding Mode=Reasoning, Model Scale=8B2026.06 | 89.1 |