Icelandic LLM - Idiom Meanings: leaderboard

The model explains in one sentence what an Icelandic idiom means, rated by an LLM judge against a reference definition as 0, 1/2 or 1.

Metric: LLM Judge Score (%). Source: huggingface.co. Saturation forecast: Around February 2027. 88 models tracked.

Top models

#ModelScoreOverall rank
1GPT-6 Astra84.61#6
2Claude Opus 584.55#16
3Claude Opus 5.584.04#3
4Claude Opus 4.883.59#39
5Claude Fable 5.183.4#10
6GPT-6.1 Sol82.82#7
7Gemini 3.7 Flash82.06#24
8Gemini 3.1 Pro (Preview)81.86#64
9Gemini 4 Argon81.67#9
10Gemini 3.8 Flash81.61#30
11GPT-6 Sol81.55#19
12Gemini 3.5 Flash81.03#66
13Claude Sonnet 580.33#65
14GPT-6.1 Terra79.76#15
15Claude Sonnet 5.579.31#14

Interactive version: theaggregate.ai/benchmark?slug=icelandic-llm-idiom-meanings · How It Works · Data refreshed daily, snapshot 2026-11-05.