EuroEval Romanian Common Sense Reasoning: leaderboard
Metric: Common Sense Reasoning Average Score (%). Source: euroeval.com. 190 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.5 (Thinking) | 80.3 |
| 2 | Gemini 3 Pro (Preview) | 79.01 |
| 3 | GPT-5 | 76.49 |
| 4 | Gemini 2.5 Pro | 72.6 |
| 5 | Claude Sonnet 4.5 (Non-reasoning) | 71.44 |
| 6 | Grok 4.1 Fast (Reasoning) | 70.12 |
| 7 | GPT-5 Mini | 69.64 |
| 8 | Gemini 2.5 Flash (Thinking) | 68.77 |
| 9 | Gemma 4 31B (IT) | 66.7 |
| 10 | Qwen 3.5 9B | 66.51 |
| 11 | GPT-4.1 | 62.13 |
| 12 | OLMo 3.1 32B (Thinking) | 58.01 |
| 13 | Gemini 3.1 Flash Lite (Preview) | 57.31 |
| 14 | Claude Sonnet 4.6 | 56.36 |
| 15 | Qwen 3.5 4B | 55.02 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-romanian-common-sense-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-05.