Evals for Every Language - Language min: leaderboard
Metric: Average Score (%). Source: huggingface.co. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 71.94 |
| 2 | Gemini 2.5 Pro | 71.1 |
| 3 | Claude Sonnet 4.5 | 70.12 |
| 4 | Claude Opus 4.8 | 69.79 |
| 5 | DeepSeek V3.1 | 69.27 |
| 6 | GPT-5.1 | 68.2 |
| 7 | Qwen 3.6 Plus | 67.66 |
| 8 | GPT-5.5 | 67.6 |
| 9 | GPT-5.3 | 67.18 |
| 10 | Claude Sonnet 4 | 67.12 |
| 11 | GPT-5.2 | 67.07 |
| 12 | Claude Opus 4.7 | 66.74 |
| 13 | DeepSeek V3.2 Exp | 66.62 |
| 14 | Gemini 3.1 Flash Lite | 66.44 |
| 15 | Claude Sonnet 4.6 | 65.74 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-min · How It Works · Data refreshed daily, snapshot 2026-09-05.