EuroEval Greek Common Sense Reasoning: leaderboard
Metric: Common Sense Reasoning Average Score (%). Source: euroeval.com. 187 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 81.52 |
| 2 | Claude Sonnet 4.5 (Thinking) | 80.87 |
| 3 | GPT-5 | 75.57 |
| 4 | Gemini 2.5 Pro | 72.39 |
| 5 | GPT-5 Mini | 69.09 |
| 6 | Gemini 2.5 Flash (Thinking) | 67.48 |
| 7 | Claude Sonnet 4.5 (Non-reasoning) | 66.68 |
| 8 | GPT-4.1 | 62.95 |
| 9 | Grok 4.1 Fast (Reasoning) | 62.73 |
| 10 | Gemma 4 31B (IT) | 62.29 |
| 11 | Qwen 3.5 9B | 60.8 |
| 12 | GPT-5.4 Mini (High) | 60.04 |
| 13 | Claude Sonnet 4.6 | 56.2 |
| 14 | Gemini 3.1 Flash Lite (Preview) | 54.05 |
| 15 | Llama 3.3 70B Instruct | 52.85 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-greek-common-sense-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-05.