EuroEval Ukrainian Common Sense Reasoning: leaderboard
EuroEval's Ukrainian common-sense reasoning task — multiple-choice questions testing everyday physical and social reasoning in Ukrainian, from the EuroEval European language model benchmark suite.
Metric: Common Sense Reasoning Average Score (%). Source: euroeval.com. Saturation forecast: Around May 2027. 185 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 82.64 | #66 |
| 2 | Claude Sonnet 4.5 (Thinking) | 81.63 | #142 (Claude Sonnet 4.5) |
| 3 | GPT-5 | 78.94 | #98 |
| 4 | Gemini 3 Flash (Preview) (Non-reasoning) | 72.33 | #82 (Gemini 3 Flash (Preview)) |
| 5 | Claude Sonnet 4.5 (Non-reasoning) | 72.24 | #142 (Claude Sonnet 4.5) |
| 6 | Gemini 2.5 Pro | 71.53 | #152 |
| 7 | Gemini 3 Flash (Preview) (Thinking) | 70.09 | #82 (Gemini 3 Flash (Preview)) |
| 8 | GPT-5 Mini (2025-08-07) | 69.68 | #172 |
| 9 | Gemini 2.5 Flash (Thinking) | 67.39 | #245 (Gemini 2.5 Flash) |
| 10 | Qwen 3.5 35B A3B Base | 64.18 | #310 |
| 11 | Gemma 4 31B (IT) | 61.68 | #176 |
| 12 | Claude Sonnet 4.6 | 60.01 | #90 |
| 13 | Qwen 3.5 9B | 58.28 | #379 |
| 14 | GPT-4.1 | 57.1 | #248 |
| 15 | Gemini 3.1 Flash Lite (Preview) | 56.95 | #194 |
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-common-sense-reasoning · How It Works · Data refreshed daily, snapshot 2026-10-20.