EuroEval Polish Knowledge: leaderboard
Metric: Knowledge Average Score (%). Source: euroeval.com. 216 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 89.17 |
| 2 | GPT-5 | 86.45 |
| 3 | Gemini 2.5 Flash (Thinking) | 85.67 |
| 4 | Gemini 2.5 Pro | 84.99 |
| 5 | GPT-5.4 Mini (High) | 83.73 |
| 6 | Claude Sonnet 4.5 (Thinking) | 82.05 |
| 7 | Claude Sonnet 4.6 | 81.99 |
| 8 | Gemini 3.1 Flash Lite (Preview) | 81.19 |
| 9 | GPT-5.4 Mini (Medium) | 80.86 |
| 10 | Grok 4.1 Fast (Reasoning) | 80.05 |
| 11 | GPT-5.4 Nano (High) | 79.77 |
| 12 | GPT-5.4 Nano (Medium) | 78.24 |
| 13 | GPT-5.2 | 77.32 |
| 14 | GPT-5 Mini | 76.52 |
| 15 | GPT-5 Nano | 70.24 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-polish-knowledge · How It Works · Data refreshed daily, snapshot 2026-09-05.