EuroEval Icelandic NLU - Hotter and Colder Sentiment: leaderboard
The "Hotter and Colder Sentiment" subtask of EuroEval Icelandic NLU. Icelandic-language EuroEval natural-language-understanding suite, separating NLU task performance from the broader all-task EuroEval aggregate.
Metric: Matthews correlation coefficient (%). Source: euroeval.com. Saturation forecast: Around January 2027. 393 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 65.21 | #66 |
| 2 | Gemini 3.1 Flash Lite (Preview) | 63.57 | #194 |
| 3 | Gemma 4 31B (IT) | 61.64 | #176 |
| 4 | GPT-5.2 | 60.78 | #110 |
| 5 | Grok 2 (1212) | 60.43 | #496 |
| 6 | Gemini 3 Flash (Preview) (Thinking) | 60.14 | #82 (Gemini 3 Flash (Preview)) |
| 7 | GPT-4.1 | 59.94 | #248 |
| 8 | Gemini 2.5 Pro (Preview 03-25) | 59.15 | #123 |
| 9 | Grok 4.1 Fast (Reasoning) | 59.12 | #217 (Grok 4.1 Fast) |
| 10 | Grok 3 | 58.9 | #304 |
| 11 | GPT-5 | 58.87 | #98 |
| 12 | Gemma 4 26B A4B (IT) | 58.78 | #297 |
| 13 | O3 (2025-04-16) | 58.74 | #121 |
| 14 | Gemini 3 Flash (Preview) (Non-reasoning) | 58.44 | #82 (Gemini 3 Flash (Preview)) |
| 15 | Gemini 2.5 Flash (Non-reasoning) | 58.32 | #245 (Gemini 2.5 Flash) |
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-nlu-hotter-and-colder-sentiment · How It Works · Data refreshed daily, snapshot 2026-10-20.