Evals for Every Language - Polish (pl): leaderboard
Average score across evaluation tasks in Polish (pl), from Evals for Every Language.
Metric: Average Score (%). Source: huggingface.co. Saturation forecast: Around February 2027. 68 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 76.56 | #58 |
| 2 | GPT-5 | 75.96 | #98 |
| 3 | Claude Sonnet 4.5 | 75.75 | #142 |
| 4 | GPT-5.2 | 75.04 | #110 |
| 5 | Gemini 2.5 Pro | 75.03 | #152 |
| 6 | Claude Opus 4.8 | 74.79 | #37 |
| 7 | Claude Opus 4.7 | 74.6 | #49 |
| 8 | Llama 4 Maverick | 74.19 | #464 |
| 9 | Claude Sonnet 4 | 73.89 | #199 |
| 10 | Claude Sonnet 4.6 | 73.79 | #90 |
| 11 | GPT-5.1 | 73.79 | #135 |
| 12 | GPT-4o | 73.44 | #343 |
| 13 | Kimi K2 | 72.69 | #243 |
| 14 | Gemini 3.1 Flash Lite | 72.66 | #191 |
| 15 | GPT-4.1 | 72.44 | #248 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=evals-for-every-language-language-pl · How It Works · Data refreshed daily, snapshot 2026-10-20.