EuroEval Icelandic — leaderboard
Metric: Average Score (%). Source: euroeval.com. 401 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 | 70.59 |
| 2 | GPT-5 (High) | 68.61 |
| 3 | Gemini 3 Pro (Preview) | 68.02 |
| 4 | O3 (2025-04-16) | 66.08 |
| 5 | Gemini 2.5 Pro | 65.81 |
| 6 | Gemini 2.5 Pro (Preview 03-25) | 64.33 |
| 7 | Gemini 2.5 Flash (Thinking) | 62.64 |
| 8 | Gemini 3 Flash (Preview) (Non-reasoning) | 61.84 |
| 9 | Gemini 2.5 Flash (Preview 04-17) | 61.64 |
| 10 | Claude 3.7 Sonnet (20250219) (Thinking) | 61.61 |
| 11 | Claude Sonnet 4.5 (Thinking) | 60.67 |
| 12 | GPT-5.4 Mini (High) | 60.51 |
| 13 | GPT-5 Mini (High) | 60.19 |
| 14 | Gemini 3.1 Flash Lite (Preview) | 59.64 |
| 15 | GPT-5.4 Mini (Medium) | 59.56 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-icelandic · How the rankings work · Data refreshed daily, snapshot 2026-07-22.