EuroEval Norwegian NLU - ScaLA NN: leaderboard
Metric: Linguistic acceptability Score (%). Source: euroeval.com. 397 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 79.31 |
| 2 | Gemini 3.1 Flash Lite (Preview) | 73.39 |
| 3 | Gemma 4 31B (IT) | 70.77 |
| 4 | O3 (2025-04-16) | 67.42 |
| 5 | GPT-5 (High) | 67.32 |
| 6 | GPT-5 | 67.11 |
| 7 | Claude Sonnet 4.6 | 66.23 |
| 8 | Gemini 2.5 Flash (Preview 04-17) | 66.16 |
| 9 | Gemini 2.5 Pro | 65.85 |
| 10 | Llama 3.1 405B | 65.37 |
| 11 | Claude 3.7 Sonnet (20250219) (Thinking) | 64.83 |
| 12 | GPT-5.4 Mini (High) | 63.61 |
| 13 | GPT-5.4 Mini (Medium) | 63.15 |
| 14 | Gemini 2.5 Pro (Preview 03-25) | 60.74 |
| 15 | GPT-5 Mini (High) | 59.92 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-norwegian-nlu-scala-nn · How It Works · Data refreshed daily, snapshot 2026-09-05.