Evals for Every Language - Language bjn — leaderboard
Metric: Average Score (%). Source: huggingface.co. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 50.24 |
| 2 | Gemini 2.5 Pro | 49.79 |
| 3 | Gemini 3.1 Flash Lite | 49.28 |
| 4 | Claude Opus 4.7 | 48.88 |
| 5 | Claude Sonnet 4.5 | 47.53 |
| 6 | Claude Opus 4.8 | 47.35 |
| 7 | Claude Sonnet 4 | 45.73 |
| 8 | Qwen 3.6 Plus | 45.34 |
| 9 | GPT-5 | 44.92 |
| 10 | Gemini 2.5 Flash Lite | 44.38 |
| 11 | GPT-5.5 | 44.06 |
| 12 | GPT-5.1 | 43.34 |
| 13 | Gemini 2.5 Flash | 43.28 |
| 14 | Mistral Medium 3.1 | 43.12 |
| 15 | Llama 3.1 70B Instruct | 42.91 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-bjn · How the rankings work · Data refreshed daily, snapshot 2026-07-22.