Icelandic LLM - ARC-Challenge-IS — leaderboard

Metric: Score (%). Source: huggingface.co. 92 models tracked.

Top models

#ModelScore
1GPT-5.595.22
2GPT-5.6 Sol95.22
3Gemini 3.1 Pro (Preview)95.14
4Gemini 3.5 Flash95.14
5Gemini 3 Pro (Preview)95.05
6Gemini 3 Flash (Preview)95.05
7Claude Opus 4.894.97
8Claude Opus 4.694.8
9GPT-5.6 Terra94.8
10Gemma 4 31B94.71
11Gemini 3.1 Flash Lite (Preview)94.62
12Grok 4.394.62
13GPT-5.4 (2026-03-05)94.62
14GPT-5.194.54
15Claude Opus 4.5 (20251101)94.54

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