Evals for Every Language - Language kn: leaderboard
Metric: Average Score (%). Source: huggingface.co. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 72.55 |
| 2 | Claude Opus 4.7 | 70.65 |
| 3 | Claude Sonnet 4 | 69.65 |
| 4 | Grok 4.20 | 69.33 |
| 5 | GPT-5.3 | 69.28 |
| 6 | GPT-5 | 68.68 |
| 7 | GPT-5.2 | 68.54 |
| 8 | Claude Sonnet 4.5 | 68.23 |
| 9 | GPT-5.1 | 68.14 |
| 10 | GPT-5.5 | 68.06 |
| 11 | Claude Sonnet 4.6 | 67.4 |
| 12 | Gemini 3.1 Flash Lite | 66.91 |
| 13 | Qwen 3.6 Plus | 66.67 |
| 14 | Claude Opus 4.8 | 66.28 |
| 15 | GPT-4o | 65.82 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-kn · How It Works · Data refreshed daily, snapshot 2026-09-05.