EuroEval Icelandic: leaderboard

Icelandic-language EuroEval all-task suite spanning NLU, knowledge, common-sense reasoning, reading comprehension, summarization, and related tasks.

Metric: Average Score (%). Source: euroeval.com. Saturation forecast: Around May 2027. 401 models tracked.

Top models

#ModelScoreOverall rank
1GPT-570.59#98
2GPT-5 (2025-08-07) (High)68.61
3Gemini 3 Pro (Preview)68.02#66
4O3 (2025-04-16)66.08#121
5Gemini 2.5 Pro65.81#152
6Gemini 2.5 Pro (Preview 03-25)64.33#123
7Gemini 3 Flash (Preview) (Thinking)63.06#82 (Gemini 3 Flash (Preview))
8Gemini 2.5 Flash (Thinking)62.64#245 (Gemini 2.5 Flash)
9Gemini 3 Flash (Preview) (Non-reasoning)61.84#82 (Gemini 3 Flash (Preview))
10Gemini 2.5 Flash (Preview 04-17)61.64#237
11Claude 3.7 Sonnet (20250219) (Thinking)61.61#201 (Claude 3.7 Sonnet (20250219))
12Claude Sonnet 4.5 (Thinking)60.67#142 (Claude Sonnet 4.5)
13GPT-5.4 Mini (2026-03-17) (High)60.51#169 (GPT-5.4 Mini (2026-03-17))
14GPT-5 Mini (2025-08-07) (High)60.19#172 (GPT-5 Mini (2025-08-07))
15Gemini 3.1 Flash Lite (Preview)59.64#194

No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.

Interactive version: theaggregate.ai/benchmark?slug=euroeval-icelandic · How It Works · Data refreshed daily, snapshot 2026-10-20.