Polish Knowledge Benchmark - Cuisine: leaderboard
Metric: Accuracy (%). Source: huggingface.co. 68 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.6 Flash | 92 |
| 2 | Gemini 3 Flash (Preview) | 91.71 |
| 3 | Gemma 4 31B (IT) | 90.29 |
| 4 | Gemini 3.1 Flash Lite | 90.29 |
| 5 | Gemini 3.1 Flash Lite (Preview) | 90 |
| 6 | Mistral Large 3 | 90 |
| 7 | Gemini 3.5 Flash Lite | 89.43 |
| 8 | Gemini 2.5 Flash | 89.14 |
| 9 | Qwen 3.5 397B A17B | 88.86 |
| 10 | DeepSeek V4 Flash | 88.57 |
| 11 | Gemma 4 26B A4B (IT) | 88 |
| 12 | Grok 4.1 Fast | 87.43 |
| 13 | Bielik-11B-v2.5-Instruct | 86.86 |
| 14 | DeepSeek V3.2 | 86.57 |
| 15 | Gemini 2.0 Flash (001) | 86.57 |
Interactive version: theaggregate.ai/benchmark?slug=polish-knowledge-benchmark-cuisine · How It Works · Data refreshed daily, snapshot 2026-09-19.