Evals for Every Language - Language ug: leaderboard
Metric: Average Score (%). Source: huggingface.co. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 70.13 |
| 2 | Claude Opus 4.7 | 68.48 |
| 3 | Claude Opus 4.8 | 67.89 |
| 4 | GPT-5.3 | 66.58 |
| 5 | GPT-5.5 | 66.4 |
| 6 | GPT-5.4 | 65.62 |
| 7 | Claude Sonnet 4.6 | 65.34 |
| 8 | Grok 4.20 | 65.25 |
| 9 | Claude Sonnet 4.5 | 65.12 |
| 10 | Claude Sonnet 4 | 63.69 |
| 11 | Gemini 3.1 Flash Lite | 63.61 |
| 12 | GPT-5.1 | 63.37 |
| 13 | Qwen 3.6 Plus | 63.12 |
| 14 | GPT-5.2 | 62.85 |
| 15 | GPT-5 | 61.76 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-ug · How It Works · Data refreshed daily, snapshot 2026-09-05.