EuroEval Finnish Common Sense Reasoning: leaderboard
Metric: Common Sense Reasoning Average Score (%). Source: euroeval.com. 338 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.5 (Non-reasoning) | 88.15 |
| 2 | Claude Sonnet 4.5 (Thinking) | 87.78 |
| 3 | GPT-5 (High) | 87.45 |
| 4 | Gemini 3 Pro (Preview) | 86.54 |
| 5 | Gemini 2.5 Pro | 86.42 |
| 6 | Claude Sonnet 4.6 | 83.76 |
| 7 | GPT-5 | 83.19 |
| 8 | Gemini 2.5 Flash (Thinking) | 82.93 |
| 9 | GPT-5 (Minimal) | 82.09 |
| 10 | Grok 3 | 80.26 |
| 11 | Gemma 4 31B (IT) | 80.17 |
| 12 | GPT-4o (2024-05-13) | 79.83 |
| 13 | Claude 3.7 Sonnet (20250219) (Thinking) | 79.46 |
| 14 | Gemini 3.1 Flash Lite (Preview) | 79.23 |
| 15 | Claude 3.5 Sonnet (20241022) | 78.49 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-finnish-common-sense-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-05.