EuroEval Norwegian NLU: leaderboard
Norwegian-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: Around March 2027. 403 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 74.92 | #66 |
| 2 | Gemma 4 31B (IT) | 74.07 | #176 |
| 3 | Gemini 3 Flash (Preview) (Non-reasoning) | 72.54 | #82 (Gemini 3 Flash (Preview)) |
| 4 | Llama 3.1 405B | 71.97 | #466 |
| 5 | GPT-5 (2025-08-07) (High) | 71.76 | |
| 6 | GPT-5 | 71.19 | #98 |
| 7 | Claude 3.7 Sonnet (20250219) (Thinking) | 71.16 | #201 (Claude 3.7 Sonnet (20250219)) |
| 8 | Gemini 3.1 Flash Lite (Preview) | 71.04 | #194 |
| 9 | GPT-5.4 Mini (2026-03-17) (High) | 70.88 | #169 (GPT-5.4 Mini (2026-03-17)) |
| 10 | Claude Sonnet 4.5 (Thinking) | 70.56 | #142 (Claude Sonnet 4.5) |
| 11 | O3 (2025-04-16) | 69.81 | #121 |
| 12 | Claude Sonnet 4.6 | 69.71 | #90 |
| 13 | Gemma 4 26B A4B (IT) | 69.67 | #297 |
| 14 | Llama 3.1 405B Instruct FP8 | 69.42 | #389 |
| 15 | Gemini 3 Flash (Preview) (Thinking) | 69.06 | #82 (Gemini 3 Flash (Preview)) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-norwegian-nlu · How It Works · Data refreshed daily, snapshot 2026-10-20.