Evals for Every Language - Language fil — leaderboard
Metric: Average Score (%). Source: huggingface.co. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 67.25 |
| 2 | GPT-4o | 66.8 |
| 3 | Gemini 2.5 Pro | 66.71 |
| 4 | Claude Sonnet 4.6 | 66.54 |
| 5 | Claude Opus 4.8 | 65.99 |
| 6 | Claude Sonnet 4 | 65.91 |
| 7 | GPT-5.1 | 65.52 |
| 8 | Claude Sonnet 4.5 | 65.34 |
| 9 | Gemini 3.1 Flash Lite | 65.09 |
| 10 | Llama 3.1 70B Instruct | 64.31 |
| 11 | DeepSeek V3.2 Exp | 64.3 |
| 12 | Kimi K2 | 63.5 |
| 13 | GPT-5.5 | 63.27 |
| 14 | Gemma 3 27B (IT) | 63.2 |
| 15 | GPT-5 | 62.81 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-fil · How the rankings work · Data refreshed daily, snapshot 2026-07-22.