Evals for Every Language - Language si: leaderboard
Metric: Average Score (%). Source: huggingface.co. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 69.9 |
| 2 | Claude Sonnet 4.5 | 66.37 |
| 3 | Qwen 3.6 Plus | 66.03 |
| 4 | Claude Opus 4.7 | 65.53 |
| 5 | Claude Sonnet 4.6 | 65.12 |
| 6 | Claude Sonnet 4 | 64.84 |
| 7 | GPT-5 | 64.76 |
| 8 | Claude Opus 4.8 | 64.48 |
| 9 | GPT-5.5 | 63.91 |
| 10 | GPT-5.3 | 63.71 |
| 11 | GPT-5.4 | 62.67 |
| 12 | GPT-5.2 | 62.61 |
| 13 | Gemini 3.1 Flash Lite | 61.55 |
| 14 | Grok 4.20 | 61.52 |
| 15 | GPT-4o | 61.05 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-si · How It Works · Data refreshed daily, snapshot 2026-09-05.