Evals for Every Language - Language kab: leaderboard
Metric: Average Score (%). Source: huggingface.co. 70 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 49.02 |
| 2 | Gemini 3.1 Flash Lite | 42.99 |
| 3 | Gemini 2.5 Pro | 42.13 |
| 4 | GPT-5.5 | 40.53 |
| 5 | Claude Opus 4.7 | 40.17 |
| 6 | Qwen 3.6 Plus | 37.9 |
| 7 | GPT-5.3 | 37.6 |
| 8 | Gemini 2.5 Flash | 37.43 |
| 9 | Claude Sonnet 4.5 | 36.29 |
| 10 | Claude Sonnet 4.6 | 34.42 |
| 11 | GPT-5.4 | 34.39 |
| 12 | Claude Opus 4.8 | 34.36 |
| 13 | Grok 4.20 | 34.22 |
| 14 | GPT-5.2 | 33.43 |
| 15 | GPT-5 | 32.97 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-kab · How It Works · Data refreshed daily, snapshot 2026-09-05.