EuroEval Croatian Common Sense Reasoning: leaderboard
Metric: Common Sense Reasoning Average Score (%). Source: euroeval.com. 194 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.5 (Thinking) | 82.86 |
| 2 | Gemini 3 Pro (Preview) | 78.16 |
| 3 | GPT-5 | 77.43 |
| 4 | Gemini 2.5 Pro | 73.33 |
| 5 | Gemini 2.5 Flash (Thinking) | 72.6 |
| 6 | Grok 4.20 Beta (0309) (Reasoning) | 72.03 |
| 7 | Qwen 3.5 9B | 66.6 |
| 8 | GPT-5 Mini | 62.76 |
| 9 | GPT-4.1 | 62.74 |
| 10 | Grok 4.1 Fast (Reasoning) | 60.41 |
| 11 | Gemma 4 31B (IT) | 59.29 |
| 12 | Claude Sonnet 4.5 (Non-reasoning) | 57.93 |
| 13 | Qwen 3.5 4B | 56.6 |
| 14 | GPT-5.4 Mini (High) | 52.97 |
| 15 | Qwen 2.5 72B Instruct | 51.52 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-croatian-common-sense-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-05.