Evals for Every Language - Language aeb: leaderboard
Metric: Average Score (%). Source: huggingface.co. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 53.74 |
| 2 | Gemini 3.1 Flash Lite | 53.18 |
| 3 | Claude Sonnet 4.5 | 51.97 |
| 4 | Claude Opus 4.7 | 50.61 |
| 5 | Claude Opus 4.8 | 50.53 |
| 6 | Gemini 2.5 Pro | 50.19 |
| 7 | Claude Sonnet 4 | 48.82 |
| 8 | Claude Sonnet 4.6 | 48.46 |
| 9 | Command A | 48.44 |
| 10 | GPT-5.2 | 47.89 |
| 11 | GPT-4o | 47.6 |
| 12 | Gemini 2.5 Flash Lite | 47.28 |
| 13 | GPT-5 | 46.56 |
| 14 | GPT-5.3 | 46.27 |
| 15 | Gemini 2.5 Flash | 45.89 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-aeb · How It Works · Data refreshed daily, snapshot 2026-09-05.