EuroEval Romanian Common Sense Reasoning: leaderboard
EuroEval's Romanian common-sense reasoning task — multiple-choice questions testing everyday physical and social reasoning in Romanian, from the EuroEval European language model benchmark suite.
Metric: Common Sense Reasoning Average Score (%). Source: euroeval.com. Saturation forecast: Around May 2027. 190 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4.5 (Thinking) | 80.3 | #142 (Claude Sonnet 4.5) |
| 2 | Gemini 3 Pro (Preview) | 79.01 | #66 |
| 3 | GPT-5 | 76.49 | #98 |
| 4 | Gemini 2.5 Pro | 72.6 | #152 |
| 5 | Claude Sonnet 4.5 (Non-reasoning) | 71.44 | #142 (Claude Sonnet 4.5) |
| 6 | Gemini 3 Flash (Preview) (Thinking) | 70.92 | #82 (Gemini 3 Flash (Preview)) |
| 7 | Grok 4.1 Fast (Reasoning) | 70.12 | #217 (Grok 4.1 Fast) |
| 8 | GPT-5 Mini (2025-08-07) | 69.64 | #172 |
| 9 | Gemini 2.5 Flash (Thinking) | 68.77 | #245 (Gemini 2.5 Flash) |
| 10 | Gemma 4 31B (IT) | 66.7 | #176 |
| 11 | Gemini 3 Flash (Preview) (Non-reasoning) | 66.52 | #82 (Gemini 3 Flash (Preview)) |
| 12 | Qwen 3.5 9B | 66.51 | #379 |
| 13 | GPT-4.1 | 62.13 | #248 |
| 14 | OLMo 3.1 32B (Thinking) | 58.01 | #632 (OLMo 3.1 32B) |
| 15 | Gemini 3.1 Flash Lite (Preview) | 57.31 | #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-romanian-common-sense-reasoning · How It Works · Data refreshed daily, snapshot 2026-10-20.