Evals for Every Language - Language szl: leaderboard
Metric: Average Score (%). Source: huggingface.co. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 | 71.4 |
| 2 | Gemini 3.1 Pro (Preview) | 71.36 |
| 3 | Claude Opus 4.8 | 69.86 |
| 4 | Claude Sonnet 4.5 | 69.85 |
| 5 | Gemini 3.1 Flash Lite | 69.81 |
| 6 | Claude Sonnet 4 | 69.44 |
| 7 | GPT-5.1 | 69.22 |
| 8 | Claude Opus 4.7 | 68.88 |
| 9 | Gemini 2.5 Pro | 68.86 |
| 10 | GPT-5.2 | 68.73 |
| 11 | DeepSeek V3.2 Exp | 68.59 |
| 12 | GPT-5.3 | 67.4 |
| 13 | Claude Sonnet 4.6 | 67.26 |
| 14 | Qwen 3.6 Plus | 67.21 |
| 15 | GPT-5.4 | 66.72 |
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-szl · How It Works · Data refreshed daily, snapshot 2026-09-05.