Evals for Every Language - Language ak: leaderboard
Metric: Average Score (%). Source: huggingface.co. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 64.43 |
| 2 | Claude Opus 4.8 | 60.02 |
| 3 | Claude Opus 4.7 | 59.75 |
| 4 | GPT-5.3 | 58.7 |
| 5 | GPT-5.5 | 57.86 |
| 6 | Gemini 2.5 Pro | 56.18 |
| 7 | Grok 4.20 | 56.11 |
| 8 | Claude Sonnet 4.5 | 56.05 |
| 9 | GPT-5.4 | 55.22 |
| 10 | Claude Sonnet 4.6 | 52.33 |
| 11 | Gemini 3.1 Flash Lite | 52.14 |
| 12 | GPT-5 | 52.1 |
| 13 | GPT-5.2 | 52.02 |
| 14 | GPT-5.1 | 51.29 |
| 15 | Qwen 3.6 Plus | 51.2 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-ak · How It Works · Data refreshed daily, snapshot 2026-09-05.