EuroEval Swedish Common Sense Reasoning: leaderboard
Metric: Common Sense Reasoning Average Score (%). Source: euroeval.com. 336 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.5 (Thinking) | 92.92 |
| 2 | Claude Sonnet 4.5 (Non-reasoning) | 90.52 |
| 3 | Qwen 3 235B A22B 2507 Instruct | 85.03 |
| 4 | Gemini 3 Pro (Preview) | 84.98 |
| 5 | Claude Sonnet 4.6 | 84.55 |
| 6 | Gemini 2.5 Pro | 82.2 |
| 7 | Mistral Small 3.1 | 80.93 |
| 8 | Gemma 4 31B (IT) | 80.01 |
| 9 | Llama 3.1 405B Instruct FP8 | 79.86 |
| 10 | Qwen 3 Next 80B A3B Instruct | 79.47 |
| 11 | Mistral Small 3 | 78.67 |
| 12 | GPT-5 (High) | 78.18 |
| 13 | GPT-5 (Minimal) | 76.76 |
| 14 | Qwen 3 Next 80B A3B (Thinking) | 76.51 |
| 15 | Gemini 3.1 Flash Lite (Preview) | 75.33 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-swedish-common-sense-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-05.