Evals for Every Language - Language brx: leaderboard
Metric: Average Score (%). Source: huggingface.co. 70 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 47.64 |
| 2 | Gemini 2.5 Pro | 46.66 |
| 3 | Claude Opus 4.7 | 42.86 |
| 4 | Gemini 3.1 Flash Lite | 42.8 |
| 5 | GPT-5.5 | 42.49 |
| 6 | Gemini 2.5 Flash | 41.73 |
| 7 | Claude Opus 4.8 | 37.4 |
| 8 | GPT-5.4 | 35.5 |
| 9 | GPT-5.3 | 34.42 |
| 10 | Gemini 2.5 Flash Lite | 34.06 |
| 11 | Claude Sonnet 4.5 | 33.77 |
| 12 | GPT-5.2 | 32.67 |
| 13 | Grok 4.20 | 32.41 |
| 14 | GPT-5 | 31.77 |
| 15 | DeepSeek V3.2 Exp | 31.56 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-brx · How It Works · Data refreshed daily, snapshot 2026-09-05.