Icelandic LLM - Generative Quality: leaderboard
The model writes a few paragraphs from a varied list of prompts and the text is scored for defects: grammar errors and out-of-vocabulary words. It penalises flawed text rather than ranking good against great.
Metric: Quality Score (%). Source: huggingface.co. Saturation forecast: Around February 2027. 88 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-6.1 Terra | 84.34 | #15 |
| 2 | GPT-6 Sol | 83.75 | #19 |
| 3 | GPT-6 Luna | 82.26 | #61 |
| 4 | GPT-6 Astra | 82.22 | #6 |
| 5 | GPT-6.1 Sol | 78.81 | #7 |
| 6 | GPT-5.6 Luna | 78.71 | #79 |
| 7 | GPT-5.6 Sol | 78.66 | #22 |
| 8 | DeepSeek V4 Pro | 75.6 | #112 |
| 9 | Claude Sonnet 5.5 | 74.93 | #14 |
| 10 | Gemini 4 Argon | 74.91 | #9 |
| 11 | GPT-5.6 Terra | 74.62 | #40 |
| 12 | Claude Opus 5.5 | 74.54 | #3 |
| 13 | Gemini 3.1 Pro (Preview) | 74.27 | #64 |
| 14 | Claude Fable 5.1 | 73.67 | #10 |
| 15 | Claude Opus 5 | 72.33 | #16 |
Interactive version: theaggregate.ai/benchmark?slug=icelandic-llm-generative-quality · How It Works · Data refreshed daily, snapshot 2026-11-05.