EuroEval Icelandic NLU - ScaLA IS — leaderboard

Metric: Linguistic acceptability Score (%). Source: euroeval.com. 398 models tracked.

Top models

#ModelScore
1Gemini 3 Pro (Preview)69.88
2Claude 3.7 Sonnet (20250219) (Thinking)62.13
3Gemini 3 Flash (Preview) (Non-reasoning)61.88
4GPT-5 (High)59.65
5Gemini 3.1 Flash Lite (Preview)59.17
6GPT-557.81
7Claude Sonnet 4.5 (Thinking)56.61
8O3 (2025-04-16)55.32
9Gemini 2.5 Flash (Thinking)53.96
10O4 Mini (2025-04-16)53.35
11Gemini 2.5 Pro53.03
12Gemini 2.5 Pro (Preview 03-25)51.46
13GPT-5 Mini (High)50.58
14GPT-4.150.04
15GPT-5.4 Mini (Medium)48.04

Interactive version: theaggregate.ai/benchmark?slug=euroeval-icelandic-nlu-scala-is · How the rankings work · Data refreshed daily, snapshot 2026-07-22.