Icelandic LLM - Belebele-IS — leaderboard
Metric: Score (%). Source: huggingface.co. 92 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 95 |
| 2 | GPT-5.5 | 94.89 |
| 3 | Gemini 3.5 Flash | 94.78 |
| 4 | Gemini 3 Pro (Preview) | 94.67 |
| 5 | GPT-5.6 Sol | 94.67 |
| 6 | Gemini 3 Flash (Preview) | 94.56 |
| 7 | Claude Opus 4.8 | 93.89 |
| 8 | DeepSeek V4 Pro | 93.89 |
| 9 | Gemini 2.5 Pro (03-25) | 93.89 |
| 10 | GPT-5.4 (2026-03-05) | 93.89 |
| 11 | Gemini 2.5 Flash | 93.78 |
| 12 | Claude Opus 4.5 (20251101) | 93.78 |
| 13 | GPT-5.2 | 93.67 |
| 14 | GPT-5.6 Terra | 93.56 |
| 15 | Claude Sonnet 4.5 | 93.44 |
Interactive version: theaggregate.ai/benchmark?slug=icelandic-llm-belebele-is · How the rankings work · Data refreshed daily, snapshot 2026-07-22.