EuroEval Ukrainian Knowledge — leaderboard
Metric: Knowledge Average Score (%). Source: euroeval.com. 185 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 91.25 |
| 2 | GPT-5 | 90.6 |
| 3 | Claude Sonnet 4.5 (Thinking) | 89.71 |
| 4 | Grok 4.1 Fast (Reasoning) | 88.88 |
| 5 | Gemini 2.5 Pro | 88.86 |
| 6 | GPT-5.4 Mini (Medium) | 87.25 |
| 7 | GPT-5 Mini | 86.73 |
| 8 | GPT-5.4 Mini (High) | 86.49 |
| 9 | Gemini 3 Flash (Preview) (Non-reasoning) | 85.96 |
| 10 | GPT-5.4 Nano (Medium) | 84.07 |
| 11 | GPT-5 Nano | 83.69 |
| 12 | Claude Sonnet 4.6 | 83.55 |
| 13 | Gemini 2.5 Flash (Thinking) | 83.54 |
| 14 | Gemini 2.5 Flash Lite (Thinking) | 82.48 |
| 15 | GPT-5.4 Nano (High) | 82.18 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-ukrainian-knowledge · How the rankings work · Data refreshed daily, snapshot 2026-07-22.