Evals for Every Language - Language dyu: leaderboard
Metric: Average Score (%). Source: huggingface.co. 70 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 35.97 |
| 2 | Claude Opus 4.7 | 33.96 |
| 3 | Gemini 3.1 Flash Lite | 31.71 |
| 4 | Gemini 2.5 Pro | 31.6 |
| 5 | GPT-5.5 | 30.43 |
| 6 | Grok 4.20 | 30.24 |
| 7 | Claude Sonnet 4.5 | 28.77 |
| 8 | Gemini 2.5 Flash | 28.53 |
| 9 | GPT-5.4 | 28.2 |
| 10 | GPT-5.1 | 27.89 |
| 11 | GPT-5 | 27.76 |
| 12 | GPT-5.3 | 27.62 |
| 13 | Claude Sonnet 4 | 27.24 |
| 14 | Claude Opus 4.8 | 26.68 |
| 15 | Qwen 3.6 Plus | 26.4 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-dyu · How It Works · Data refreshed daily, snapshot 2026-09-05.