Polish Knowledge Benchmark - Economy & Business: leaderboard
Topic category of the Polish Knowledge Benchmark (PolishQA) covering the Polish economy and business. The benchmark has 15,000 synthetic four-option questions about Poland generated by DeepSeek 3.2 and difficulty-rated by Qwen3.5-397B-A17B. Accuracy is correct over total in the category, 0-100, higher is better.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Around February 2027. 69 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.6 Flash | 85.85 | #57 |
| 2 | Gemini 3 Flash (Preview) | 85.69 | #90 |
| 3 | Mistral Large 4 | 84.39 | #106 |
| 4 | Gemini 3.1 Flash Lite | 84.23 | #200 |
| 5 | Gemini 3.1 Flash Lite (Preview) | 83.9 | #201 |
| 6 | DeepSeek V4 Flash | 83.74 | #155 |
| 7 | Qwen 3.5 397B A17B | 82.76 | #154 |
| 8 | Gemini 2.5 Flash | 82.44 | #255 |
| 9 | Gemini 3.5 Flash Lite | 82.36 | #174 |
| 10 | Mistral Large 3 | 81.71 | #410 |
| 11 | Grok 4.1 Fast | 81.46 | #227 |
| 12 | Gemma 4 31B (IT) | 81.06 | #178 |
| 13 | Mistral Medium 3.1 | 80.89 | #418 |
| 14 | DeepSeek V3.2 | 80.16 | #210 |
| 15 | Llama 4 Maverick | 80.08 | #486 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=polish-knowledge-benchmark-economy-business · How It Works · Data refreshed daily, snapshot 2026-11-05.