Evals for Every Language - Chinese (zh): leaderboard
Average score across evaluation tasks in Chinese (zh), from Evals for Every Language.
Metric: Average Score (%). Source: huggingface.co. Saturation forecast: Around May 2027. 68 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 77.39 | #58 |
| 2 | Claude Sonnet 4.5 | 76.47 | #142 |
| 3 | Claude Sonnet 4 | 76.26 | #199 |
| 4 | Command A | 75.55 | #508 |
| 5 | GPT-5 | 75.48 | #98 |
| 6 | Claude Sonnet 4.6 | 75.34 | #90 |
| 7 | GPT-5.2 | 75.27 | #110 |
| 8 | Grok 4.20 | 75.02 | #175 |
| 9 | Gemini 3.1 Flash Lite | 74.92 | #191 |
| 10 | GPT-5.3 | 74.87 | #103 |
| 11 | jamba-large-1.7 | 74.82 | #525 |
| 12 | Mistral Medium 3.1 | 74.77 | #408 |
| 13 | Gemini 2.5 Pro | 74.61 | #152 |
| 14 | GPT-4o | 74.02 | #343 |
| 15 | Kimi K2 | 73.1 | #243 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-zh · How It Works · Data refreshed daily, snapshot 2026-10-20.