EuroEval Finnish Common Sense Reasoning: leaderboard
EuroEval's Finnish common-sense reasoning task — multiple-choice questions testing everyday physical and social reasoning in Finnish, from the EuroEval European language model benchmark suite.
Metric: Common Sense Reasoning Average Score (%). Source: euroeval.com. Saturation forecast: Around May 2027. 338 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Flash (Preview) (Thinking) | 89.66 | #82 (Gemini 3 Flash (Preview)) |
| 2 | Claude Sonnet 4.5 (Non-reasoning) | 88.15 | #142 (Claude Sonnet 4.5) |
| 3 | Claude Sonnet 4.5 (Thinking) | 87.78 | #142 (Claude Sonnet 4.5) |
| 4 | GPT-5 (2025-08-07) (High) | 87.45 | |
| 5 | Gemini 3 Flash (Preview) (Non-reasoning) | 87.08 | #82 (Gemini 3 Flash (Preview)) |
| 6 | Gemini 3 Pro (Preview) | 86.54 | #66 |
| 7 | Gemini 2.5 Pro | 86.42 | #152 |
| 8 | Claude Sonnet 4.6 | 83.76 | #90 |
| 9 | GPT-5 | 83.19 | #98 |
| 10 | Gemini 2.5 Flash (Thinking) | 82.93 | #245 (Gemini 2.5 Flash) |
| 11 | Qwen 3.5 35B A3B FP8 | 81.71 | #362 |
| 12 | Grok 3 | 80.26 | #304 |
| 13 | Gemma 4 31B (IT) | 80.17 | #176 |
| 14 | GPT-4o (2024-05-13) | 79.83 | #325 |
| 15 | Claude 3.7 Sonnet (20250219) (Thinking) | 79.46 | #201 (Claude 3.7 Sonnet (20250219)) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-finnish-common-sense-reasoning · How It Works · Data refreshed daily, snapshot 2026-10-20.