Evals for Every Language - Language doi: leaderboard
Metric: Average Score (%). Source: huggingface.co. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.7 | 71.84 |
| 2 | GPT-5.5 | 71.32 |
| 3 | Claude Opus 4.8 | 70.16 |
| 4 | Gemini 3.1 Pro (Preview) | 69.74 |
| 5 | Claude Sonnet 4.5 | 69.56 |
| 6 | GPT-5.4 | 68.61 |
| 7 | Claude Sonnet 4.6 | 68.57 |
| 8 | Claude Sonnet 4 | 68.51 |
| 9 | Gemini 2.5 Pro | 68.46 |
| 10 | GPT-5.3 | 67.17 |
| 11 | GPT-5 | 66.78 |
| 12 | GPT-5.1 | 66.56 |
| 13 | Gemini 3.1 Flash Lite | 65.25 |
| 14 | GPT-5.2 | 64.95 |
| 15 | Grok 4.20 | 63.79 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-doi · How It Works · Data refreshed daily, snapshot 2026-09-05.