Evals for Every Language - Language mn: leaderboard
Metric: Average Score (%). Source: huggingface.co. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 72.36 |
| 2 | GPT-5.2 | 70.7 |
| 3 | Claude Sonnet 4.5 | 70.62 |
| 4 | Claude Sonnet 4.6 | 70.38 |
| 5 | Claude Opus 4.7 | 70.38 |
| 6 | Gemini 3.1 Flash Lite | 69.48 |
| 7 | GPT-5 | 68.8 |
| 8 | GPT-5.1 | 68.45 |
| 9 | GPT-5.5 | 68.42 |
| 10 | Qwen 3.6 Plus | 67.84 |
| 11 | Claude Sonnet 4 | 67.76 |
| 12 | Claude Opus 4.8 | 66.93 |
| 13 | GPT-5.3 | 66.85 |
| 14 | Grok 4.20 | 66.06 |
| 15 | Qwen 3.6 Flash | 65.19 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-mn · How It Works · Data refreshed daily, snapshot 2026-09-05.