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
3GPT-5 (High)59.65
4Gemini 3.1 Flash Lite (Preview)59.17
5GPT-557.81
6Claude Sonnet 4.5 (Thinking)56.61
7O3 (2025-04-16)55.32
8Gemini 2.5 Flash (Thinking)53.96
9O4 Mini (2025-04-16)53.35
10Gemini 2.5 Pro53.03
11Gemini 2.5 Pro (Preview 03-25)51.46
12GPT-5 Mini (High)50.58
13GPT-4.150.04
14GPT-5.4 Mini (Medium)48.04
15GPT-5.247.68

Interactive version: theaggregate.ai/benchmark?slug=euroeval-icelandic-nlu-scala-is · How It Works · Data refreshed daily, snapshot 2026-09-05.