Polish Knowledge Benchmark - History: leaderboard
Metric: Accuracy (%). Source: huggingface.co. 68 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Flash (Preview) | 94.41 |
| 2 | Gemini 3.6 Flash | 93.74 |
| 3 | DeepSeek V4 Flash | 93.06 |
| 4 | Gemini 3.1 Flash Lite (Preview) | 92.47 |
| 5 | Gemini 3.1 Flash Lite | 92.39 |
| 6 | Mistral Large 3 | 91.99 |
| 7 | Qwen 3.5 397B A17B | 91.48 |
| 8 | Grok 4.1 Fast | 91.12 |
| 9 | DeepSeek V3.2 | 90.92 |
| 10 | Gemini 2.5 Flash | 90.84 |
| 11 | Gemini 3.5 Flash Lite | 90.64 |
| 12 | Gemini 2.0 Flash (001) | 90.33 |
| 13 | Mistral Medium 3.1 | 90.21 |
| 14 | DeepSeek V3 (0324) | 90.05 |
| 15 | Gemma 4 31B (IT) | 89.89 |
Interactive version: theaggregate.ai/benchmark?slug=polish-knowledge-benchmark-history · How It Works · Data refreshed daily, snapshot 2026-09-19.