Evals for Every Language - Language kam — leaderboard
Metric: Average Score (%). Source: huggingface.co. 70 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 40.75 |
| 2 | Gemini 3.1 Flash Lite | 40.59 |
| 3 | Claude Opus 4.7 | 40.02 |
| 4 | Gemini 2.5 Pro | 39.84 |
| 5 | Claude Sonnet 4.5 | 39.17 |
| 6 | GPT-5.1 | 36.02 |
| 7 | GPT-5 | 35.54 |
| 8 | Gemini 2.5 Flash | 34.97 |
| 9 | Grok 4.20 | 34.9 |
| 10 | GPT-5.5 | 34.87 |
| 11 | Mistral Medium 3.1 | 34.32 |
| 12 | Claude Opus 4.8 | 32.93 |
| 13 | GPT-4o | 32.92 |
| 14 | Llama 4 Maverick | 31.75 |
| 15 | Claude Sonnet 4.6 | 31.24 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-kam · How the rankings work · Data refreshed daily, snapshot 2026-07-22.