Evals for Every Language - Language ks: leaderboard
Metric: Average Score (%). Source: huggingface.co. 70 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 | 45.36 |
| 2 | Claude Sonnet 4.5 | 45.3 |
| 3 | Gemini 2.5 Pro | 44.55 |
| 4 | Gemini 3.1 Pro (Preview) | 44.06 |
| 5 | Gemini 2.5 Flash | 43.89 |
| 6 | Claude Opus 4.7 | 41.93 |
| 7 | Claude Sonnet 4 | 41.93 |
| 8 | GPT-5.3 | 41.76 |
| 9 | GPT-5.1 | 40.16 |
| 10 | Grok 4.20 | 39.56 |
| 11 | Gemini 2.5 Flash Lite | 39.39 |
| 12 | GPT-5 | 38.76 |
| 13 | Qwen 3.6 Plus | 38.62 |
| 14 | GPT-5.2 | 37.92 |
| 15 | Gemini 3.1 Flash Lite | 37.88 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-ks · How It Works · Data refreshed daily, snapshot 2026-09-05.