Polish Knowledge Benchmark - Literature & Polish Language: leaderboard
Metric: Accuracy (%). Source: huggingface.co. 68 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Flash (Preview) | 97.12 |
| 2 | Gemini 3.6 Flash | 94.95 |
| 3 | Gemini 3.1 Flash Lite (Preview) | 93.27 |
| 4 | Gemini 3.1 Flash Lite | 93.27 |
| 5 | Gemini 3.5 Flash Lite | 92.31 |
| 6 | DeepSeek V4 Flash | 92.07 |
| 7 | DeepSeek V3.2 | 91.83 |
| 8 | Gemini 2.0 Flash (001) | 91.11 |
| 9 | DeepSeek V3 (0324) | 90.62 |
| 10 | Mistral Large 3 | 90.62 |
| 11 | Mistral Medium 3.1 | 89.18 |
| 12 | Grok 4.1 Fast | 89.18 |
| 13 | Gemini 2.5 Flash | 88.7 |
| 14 | Qwen 3.5 397B A17B | 88.46 |
| 15 | Llama 4 Maverick | 88.22 |
Interactive version: theaggregate.ai/benchmark?slug=polish-knowledge-benchmark-literature-polish-language · How It Works · Data refreshed daily, snapshot 2026-09-19.