Evals for Every Language - Language mni — leaderboard
Metric: Average Score (%). Source: huggingface.co. 70 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 49.55 |
| 2 | Gemini 2.5 Pro | 48.21 |
| 3 | GPT-5.5 | 43.54 |
| 4 | Claude Opus 4.8 | 41.87 |
| 5 | Claude Opus 4.7 | 41.68 |
| 6 | Gemini 3.1 Flash Lite | 41.07 |
| 7 | GPT-5.4 | 39.45 |
| 8 | Gemini 2.5 Flash | 39.43 |
| 9 | Qwen 3.6 Plus | 36.05 |
| 10 | Claude Sonnet 4 | 35.27 |
| 11 | Claude Sonnet 4.6 | 34.97 |
| 12 | Claude Sonnet 4.5 | 33.33 |
| 13 | Grok 4.20 | 33.2 |
| 14 | Gemini 2.5 Flash Lite | 32.86 |
| 15 | GPT-5 | 32.32 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-mni · How the rankings work · Data refreshed daily, snapshot 2026-07-22.