EuroEval Albanian NLU - MMS SQ — leaderboard
Metric: Sentiment classification Score (%). Source: euroeval.com. 201 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Flash (Preview) (Non-reasoning) | 32.13 |
| 2 | Mistral Small 3.1 | 30.53 |
| 3 | Claude Sonnet 4.6 | 30.21 |
| 4 | Mistral Small 3.2 | 29.62 |
| 5 | Gemini 3.1 Pro (Preview) | 29.46 |
| 6 | Gemma 3 27B (IT) | 29.39 |
| 7 | GPT-5.4 Mini (High) | 29.39 |
| 8 | GPT-5.4 Nano (High) | 29.38 |
| 9 | Gemini 2.5 Pro | 28.98 |
| 10 | GPT-5.2 | 28.93 |
| 11 | Qwen 3 30B A3B 2507 Instruct | 28.48 |
| 12 | Gemini 2.5 Flash (Non-reasoning) | 28.48 |
| 13 | Gemini 3.1 Flash Lite (Preview) | 28.43 |
| 14 | GPT-OSS-20B (Medium) | 27.91 |
| 15 | Qwen 3 14B (Non-reasoning) | 27.82 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-albanian-nlu-mms-sq · How the rankings work · Data refreshed daily, snapshot 2026-07-22.