Evals for Every Language - Language an — leaderboard
Metric: Average Score (%). Source: huggingface.co. 70 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.5 | 56.84 |
| 2 | Gemini 2.5 Pro | 56.08 |
| 3 | Gemini 3.1 Pro (Preview) | 55.86 |
| 4 | Claude Sonnet 4 | 54.78 |
| 5 | Claude Opus 4.7 | 54.06 |
| 6 | Claude Opus 4.8 | 54.04 |
| 7 | GPT-5 | 53.99 |
| 8 | Gemma 3 27B (IT) | 52.56 |
| 9 | Gemini 3.1 Flash Lite | 52.49 |
| 10 | Llama 4 Maverick | 52.46 |
| 11 | Claude Sonnet 4.6 | 52.21 |
| 12 | Nova Pro (v1) | 52.2 |
| 13 | GPT-4o | 51.9 |
| 14 | Gemini 2.5 Flash | 51.59 |
| 15 | GPT-5.4 | 51.45 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-an · How the rankings work · Data refreshed daily, snapshot 2026-07-22.