EuroEval Swedish NLU: leaderboard
Swedish-language EuroEval natural-language-understanding suite, separating NLU task performance from the broader all-task EuroEval aggregate.
Metric: NLU Average Score (%). Source: euroeval.com. Saturation forecast: Estimated already saturated. 412 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude 3.7 Sonnet (20250219) (Thinking) | 77.77 | #201 (Claude 3.7 Sonnet (20250219)) |
| 2 | Gemma 4 31B (IT) | 76.62 | #176 |
| 3 | Gemini 3 Pro (Preview) | 76.3 | #66 |
| 4 | GPT-5 (2025-08-07) (High) | 75.7 | |
| 5 | Llama 3.1 70B | 75.55 | #592 |
| 6 | Llama 3.1 405B Instruct FP8 | 75.35 | #389 |
| 7 | Llama 3.1 405B | 75.31 | #466 |
| 8 | Claude Sonnet 4.5 (Thinking) | 74.52 | #142 (Claude Sonnet 4.5) |
| 9 | Claude Sonnet 4.6 | 74.4 | #90 |
| 10 | GPT-5.4 Mini (2026-03-17) (High) | 74.39 | #169 (GPT-5.4 Mini (2026-03-17)) |
| 11 | GPT-5.2 | 74.1 | #110 |
| 12 | GPT-5 | 74.06 | #98 |
| 13 | Gemma 4 26B A4B (IT) | 74.05 | #297 |
| 14 | Mistral Small 3.1 | 73.88 | #610 |
| 15 | Gemini 2.5 Flash (Thinking) | 73.5 | #245 (Gemini 2.5 Flash) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-swedish-nlu · How It Works · Data refreshed daily, snapshot 2026-10-20.