Evals for Every Language - Language km: leaderboard
Metric: Average Score (%). Source: huggingface.co. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 68.89 |
| 2 | Claude Sonnet 4.6 | 68.03 |
| 3 | Claude Opus 4.7 | 67.86 |
| 4 | Claude Sonnet 4.5 | 66.83 |
| 5 | Claude Sonnet 4 | 66.25 |
| 6 | GPT-5.3 | 65.49 |
| 7 | Claude Opus 4.8 | 64.92 |
| 8 | GPT-5.5 | 64.78 |
| 9 | Qwen 3.6 Plus | 64.53 |
| 10 | DeepSeek V3.1 | 63.16 |
| 11 | GPT-5 | 63.14 |
| 12 | GPT-5.1 | 62.67 |
| 13 | Llama 4 Maverick | 62.46 |
| 14 | GPT-5.4 | 62.08 |
| 15 | Grok 4.20 | 61.9 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-km · How It Works · Data refreshed daily, snapshot 2026-09-05.