Polish Knowledge Benchmark - Overall: leaderboard
Metric: Accuracy (%). Source: huggingface.co. 68 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Flash (Preview) | 92.62 |
| 2 | Gemini 3.6 Flash | 91.38 |
| 3 | Gemini 3.1 Flash Lite (Preview) | 90.07 |
| 4 | Gemini 3.1 Flash Lite | 89.92 |
| 5 | DeepSeek V4 Flash | 88.91 |
| 6 | Gemini 3.5 Flash Lite | 88.37 |
| 7 | Qwen 3.5 397B A17B | 87.16 |
| 8 | Gemini 2.5 Flash | 87.15 |
| 9 | Mistral Large 3 | 87.09 |
| 10 | Grok 4.1 Fast | 86.33 |
| 11 | DeepSeek V3.2 | 86.06 |
| 12 | Mistral Medium 3.1 | 85.69 |
| 13 | Gemini 2.0 Flash (001) | 85.41 |
| 14 | Gemma 4 31B (IT) | 84.73 |
| 15 | DeepSeek V3 (0324) | 84.11 |
Interactive version: theaggregate.ai/benchmark?slug=polish-knowledge-benchmark-overall · How It Works · Data refreshed daily, snapshot 2026-09-19.