Evals for Every Language - Language zgh: leaderboard
Metric: Average Score (%). Source: huggingface.co. 70 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.6 | 46.84 |
| 2 | Gemini 2.5 Pro | 41 |
| 3 | GPT-5.5 | 39.49 |
| 4 | Gemini 2.5 Flash | 36.01 |
| 5 | Claude Opus 4.8 | 34.61 |
| 6 | Claude Opus 4.7 | 34.43 |
| 7 | Gemini 3.1 Pro (Preview) | 34.26 |
| 8 | GPT-5.3 | 34.01 |
| 9 | Gemini 3.1 Flash Lite | 33.46 |
| 10 | GPT-5.4 | 31.47 |
| 11 | Llama 4 Maverick | 29.4 |
| 12 | Qwen 3.6 Plus | 29.4 |
| 13 | Claude Sonnet 4.5 | 29.2 |
| 14 | GPT-5.2 | 26.34 |
| 15 | Grok 4.20 | 25.78 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-zgh · How It Works · Data refreshed daily, snapshot 2026-09-05.