EuroEval Swedish NLU - Multi Wiki QA SV — leaderboard

Metric: Reading comprehension Score (%). Source: euroeval.com. 254 models tracked.

Top models

#ModelScore
1Magistral Small82.4
2Mistral Small 3.281.73
3gemma-3-27B-pt81.32
4Llama 3.1 70B81.23
5SOLAR-10.7B-v1.080.55
6Ministral-3-14B-Reasoning-251280.5
7Mistral Small 3.180.08
8Qwen 3 14B (Non-reasoning)80.03
9Bielik-11B-v2.3-Instruct79.79
10Ministral 3 14B79.48
11Qwen 2.5 72B Instruct79.22
12Gemma 3 27B (IT)77.61
13gemma-3-12B-pt77.22
14Claude Sonnet 4.5 (Non-reasoning)76.41
15Llama 3.1 8B76.34

Interactive version: theaggregate.ai/benchmark?slug=euroeval-swedish-nlu-multi-wiki-qa-sv · How the rankings work · Data refreshed daily, snapshot 2026-07-22.