Evals for Every Language - Language ceb: leaderboard
Metric: Average Score (%). Source: huggingface.co. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Flash Lite | 78.06 |
| 2 | Gemini 3.1 Pro (Preview) | 77.77 |
| 3 | Claude Sonnet 4.5 | 76.66 |
| 4 | GPT-5.1 | 76 |
| 5 | Claude Opus 4.8 | 75.82 |
| 6 | Gemini 2.5 Pro | 75.18 |
| 7 | Claude Opus 4.7 | 75.18 |
| 8 | Grok 4.20 | 74.99 |
| 9 | GPT-5.2 | 74.67 |
| 10 | GPT-5.5 | 74.54 |
| 11 | Claude Sonnet 4 | 74.07 |
| 12 | GPT-5.3 | 74.06 |
| 13 | GPT-4o | 73.71 |
| 14 | GPT-5.4 | 73.65 |
| 15 | Llama 4 Maverick | 72.93 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-ceb · How It Works · Data refreshed daily, snapshot 2026-09-05.