EuroEval Norwegian Knowledge - Idioms NO: leaderboard
Metric: MCC (x100). Source: euroeval.com. 426 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 87.33 |
| 2 | Gemini 2.5 Pro (Preview 03-25) | 87.27 |
| 3 | Gemini 3 Pro (Preview) | 86.22 |
| 4 | O3 (2025-04-16) | 84.93 |
| 5 | GPT-5 (High) | 84.08 |
| 6 | GPT-5 | 83.91 |
| 7 | Gemini 3 Flash (Preview) (Non-reasoning) | 78.16 |
| 8 | GPT-5.4 Mini (High) | 75.06 |
| 9 | Claude Sonnet 4.5 (Thinking) | 73.92 |
| 10 | GPT-5.4 Mini (Medium) | 72.59 |
| 11 | GPT-5 Mini (High) | 69.06 |
| 12 | GPT-5 Mini | 66.84 |
| 13 | O4 Mini (2025-04-16) | 66.12 |
| 14 | Gemini 2.5 Flash (Preview 04-17) | 65.02 |
| 15 | Claude 3.7 Sonnet (20250219) (Thinking) | 64.28 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-norwegian-knowledge-idioms-no · How It Works · Data refreshed daily, snapshot 2026-09-19.