Evals for Every Language - Language zu — leaderboard
Metric: Average Score (%). Source: huggingface.co. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 72.93 |
| 2 | Grok 4.20 | 70.06 |
| 3 | Claude Opus 4.7 | 69.52 |
| 4 | Claude Sonnet 4.6 | 69.46 |
| 5 | GPT-5.2 | 68.89 |
| 6 | Claude Opus 4.8 | 67.44 |
| 7 | GPT-5.4 | 67.24 |
| 8 | Gemini 3.1 Flash Lite | 66.98 |
| 9 | Qwen 3.6 Plus | 66.95 |
| 10 | GPT-5.5 | 66.94 |
| 11 | DeepSeek V3.1 | 64.27 |
| 12 | GPT-5.1 | 61.92 |
| 13 | GPT-5 | 61.12 |
| 14 | Claude Sonnet 4 | 60.55 |
| 15 | Claude Sonnet 4.5 | 60.29 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-zu · How the rankings work · Data refreshed daily, snapshot 2026-07-22.