EuroEval German Common Sense Reasoning: leaderboard
EuroEval's German common-sense reasoning task — multiple-choice questions testing everyday physical and social reasoning in German, from the EuroEval European language model benchmark suite.
Metric: Common Sense Reasoning Average Score (%). Source: euroeval.com. Saturation forecast: Around May 2027. 310 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4.5 (Non-reasoning) | 92.63 | #142 (Claude Sonnet 4.5) |
| 2 | Claude Sonnet 4.5 (Thinking) | 90.85 | #142 (Claude Sonnet 4.5) |
| 3 | Qwen 3 235B A22B 2507 Instruct | 86.72 | #300 |
| 4 | Mistral Small 3.1 | 84.1 | #610 |
| 5 | Qwen 3 235B A22B 2507 FP8 (Thinking) | 82.99 | #246 (Qwen 3 235B A22B 2507 FP8) |
| 6 | Gemini 3 Flash (Preview) (Non-reasoning) | 82.82 | #82 (Gemini 3 Flash (Preview)) |
| 7 | Llama 3.1 405B Instruct FP8 | 82.81 | #389 |
| 8 | Claude Sonnet 4.6 | 82.68 | #90 |
| 9 | Gemini 3 Pro (Preview) | 82.16 | #66 |
| 10 | Mistral Small 3 | 82.04 | #627 |
| 11 | Qwen 3 Next 80B A3B Instruct | 81.74 | #368 |
| 12 | Gemini 3 Flash (Preview) (Thinking) | 81.19 | #82 (Gemini 3 Flash (Preview)) |
| 13 | GPT-5 (2025-08-07) (High) | 81.13 | |
| 14 | Gemini 2.5 Pro | 81.11 | #152 |
| 15 | GPT-5 | 79.73 | #98 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-german-common-sense-reasoning · How It Works · Data refreshed daily, snapshot 2026-10-20.