EuroEval Albanian NLU: leaderboard
Albanian-language EuroEval natural-language-understanding suite, separating NLU task performance from the broader all-task EuroEval aggregate.
Metric: NLU Average Score (%). Source: euroeval.com. Saturation forecast: Around May 2027. 209 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 61.17 | #58 |
| 2 | Gemini 3 Pro (Preview) | 59.09 | #66 |
| 3 | Gemini 3 Flash (Preview) (Non-reasoning) | 56.31 | #82 (Gemini 3 Flash (Preview)) |
| 4 | GPT-5.4 Mini (2026-03-17) (High) | 54.57 | #169 (GPT-5.4 Mini (2026-03-17)) |
| 5 | Gemma 4 31B (IT) | 54.55 | #176 |
| 6 | Gemini 3.1 Flash Lite (Preview) | 54.54 | #194 |
| 7 | Gemini 3 Flash (Preview) (Thinking) | 54.36 | #82 (Gemini 3 Flash (Preview)) |
| 8 | Mistral Small 3.2 | 53.56 | #605 |
| 9 | Gemini 2.5 Pro | 53 | #152 |
| 10 | GPT-4.1 | 52.93 | #248 |
| 11 | Gemma 3 27B (IT) | 52.49 | #523 |
| 12 | Mistral Small 3.1 | 52.19 | #610 |
| 13 | gemma-3-27B-pt | 52.09 | #502 |
| 14 | Claude Sonnet 4.5 (Thinking) | 52.02 | #142 (Claude Sonnet 4.5) |
| 15 | Gemini 2.5 Flash (Non-reasoning) | 51.78 | #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-albanian-nlu · How It Works · Data refreshed daily, snapshot 2026-10-20.