Evals for Every Language - Language en: leaderboard
Metric: Average Score (%). Source: huggingface.co. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Flash Lite | 87.28 |
| 2 | Claude Opus 4.8 | 86.15 |
| 3 | Gemini 3.1 Pro (Preview) | 85.6 |
| 4 | GPT-5.5 | 85.03 |
| 5 | Claude Opus 4.7 | 84.79 |
| 6 | Claude Sonnet 4.6 | 84.4 |
| 7 | Grok 4.20 | 84.29 |
| 8 | GPT-5.3 | 83.33 |
| 9 | Qwen 3.6 Plus | 82.75 |
| 10 | GPT-5.2 | 81.79 |
| 11 | nova-2-lite-v1 | 81.54 |
| 12 | GPT-5.4 | 81.52 |
| 13 | DeepSeek V3.1 | 80.49 |
| 14 | Qwen 3 30B A3B | 79.93 |
| 15 | Qwen 3.6 Flash | 79.89 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-en · How It Works · Data refreshed daily, snapshot 2026-09-05.