EuroEval Swedish NLU - ScaLA SV: leaderboard
The "ScaLA SV" subtask of EuroEval Swedish NLU. Swedish-language EuroEval natural-language-understanding suite, separating NLU task performance from the broader all-task EuroEval aggregate.
Metric: Matthews correlation coefficient (%). Source: euroeval.com. Saturation forecast: Estimated already saturated. 407 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude 3.7 Sonnet (20250219) (Thinking) | 80.86 | #201 (Claude 3.7 Sonnet (20250219)) |
| 2 | Gemini 3 Pro (Preview) | 78.42 | #66 |
| 3 | Gemini 2.5 Flash (Thinking) | 75.22 | #245 (Gemini 2.5 Flash) |
| 4 | Gemini 3.1 Flash Lite (Preview) | 74.34 | #194 |
| 5 | Gemini 3 Flash (Preview) (Non-reasoning) | 73.05 | #82 (Gemini 3 Flash (Preview)) |
| 6 | Gemini 2.5 Pro (Preview 03-25) | 72.68 | #123 |
| 7 | Llama 3.1 405B | 72.36 | #466 |
| 8 | O3 (2025-04-16) | 71.99 | #121 |
| 9 | Grok 3 | 71.91 | #304 |
| 10 | Gemma 4 31B (IT) | 71.8 | #176 |
| 11 | Gemini 2.5 Flash (Preview 04-17) | 71.74 | #237 |
| 12 | Gemini 2.5 Pro | 71.37 | #152 |
| 13 | GPT-5 (2025-08-07) (High) | 71.37 | |
| 14 | Claude Sonnet 4.6 | 71.29 | #90 |
| 15 | GPT-5 | 71.05 | #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-swedish-nlu-scala-sv · How It Works · Data refreshed daily, snapshot 2026-10-20.