Evals for Every Language - Language kac: leaderboard
Metric: Average Score (%). Source: huggingface.co. 70 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 42.92 |
| 2 | Claude Opus 4.7 | 39.86 |
| 3 | Gemini 2.5 Pro | 37.79 |
| 4 | GPT-5.5 | 36.93 |
| 5 | Gemini 3.1 Flash Lite | 36.66 |
| 6 | Claude Opus 4.8 | 36.24 |
| 7 | Grok 4.20 | 34.56 |
| 8 | GPT-5.3 | 34.41 |
| 9 | GPT-5.4 | 34.4 |
| 10 | DeepSeek V3.2 Exp | 32.9 |
| 11 | Kimi K2 | 32.37 |
| 12 | Gemini 2.5 Flash | 31.98 |
| 13 | Claude Sonnet 4.5 | 30.27 |
| 14 | Claude Sonnet 4.6 | 29.46 |
| 15 | DeepSeek V3.1 | 29.46 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-kac · How It Works · Data refreshed daily, snapshot 2026-09-05.