EuroEval Ukrainian NLU: leaderboard
Ukrainian-language EuroEval natural-language-understanding suite, separating NLU task performance from the broader all-task EuroEval aggregate.
Metric: NLU Average Score (%). Source: euroeval.com. Saturation forecast: Estimated already saturated. 205 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Mistral Small 3.2 | 60.27 | #605 |
| 2 | gemma-3-27B-pt | 59.95 | #502 |
| 3 | Gemma 4 31B (IT) | 59.49 | #176 |
| 4 | GPT-5.4 Mini (2026-03-17) (High) | 59.24 | #169 (GPT-5.4 Mini (2026-03-17)) |
| 5 | Voxtral-Small-24B-2507 | 58.95 | #581 |
| 6 | Gemini 3 Pro (Preview) | 58.89 | #66 |
| 7 | Mistral Small 3.1 | 58.88 | #610 |
| 8 | Ministral-3-14B-Reasoning-2512 | 58.81 | #540 |
| 9 | GPT-5 | 58.58 | #98 |
| 10 | Gemini 3.1 Flash Lite (Preview) | 58.44 | #194 |
| 11 | Claude Sonnet 4.5 (Thinking) | 58.36 | #142 (Claude Sonnet 4.5) |
| 12 | Gemini 2.5 Flash (Thinking) | 58.1 | #245 (Gemini 2.5 Flash) |
| 13 | Llama 3.1 70B | 57.84 | #592 |
| 14 | Llama 3.3 70B Instruct | 57.23 | #534 |
| 15 | Gemini 2.5 Flash Lite (Thinking) | 57.09 | #422 (Gemini 2.5 Flash Lite) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-ukrainian-nlu · How It Works · Data refreshed daily, snapshot 2026-10-20.