Evals for Every Language - Language fo: leaderboard
Metric: Average Score (%). Source: huggingface.co. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 63.87 |
| 2 | Gemini 2.5 Flash | 61.13 |
| 3 | Gemini 3.1 Pro (Preview) | 59.62 |
| 4 | GPT-5.3 | 57.71 |
| 5 | GPT-5 | 56.28 |
| 6 | Claude Sonnet 4.5 | 55.87 |
| 7 | GPT-4o | 55.85 |
| 8 | Gemini 3.1 Flash Lite | 54.79 |
| 9 | Claude Sonnet 4 | 54.4 |
| 10 | GPT-5.5 | 54.38 |
| 11 | Gemini 2.5 Flash Lite | 53.67 |
| 12 | Qwen 3.6 Plus | 53.49 |
| 13 | GPT-5.1 | 51.95 |
| 14 | GPT-5.2 | 51.89 |
| 15 | Claude Opus 4.7 | 51.5 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-fo · How It Works · Data refreshed daily, snapshot 2026-09-05.