EuroEval Albanian NLU - MMS SQ: leaderboard
The "MMS SQ" subtask of EuroEval Albanian NLU. Albanian-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 2030. 201 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Flash (Preview) (Non-reasoning) | 32.13 | #82 (Gemini 3 Flash (Preview)) |
| 2 | Mistral Small 3.1 | 30.53 | #610 |
| 3 | Claude Sonnet 4.6 | 30.21 | #90 |
| 4 | Mistral Small 3.2 | 29.62 | #605 |
| 5 | Gemini 3.1 Pro (Preview) | 29.46 | #58 |
| 6 | Gemma 3 27B (IT) | 29.39 | #523 |
| 7 | GPT-5.4 Mini (2026-03-17) (High) | 29.39 | #169 (GPT-5.4 Mini (2026-03-17)) |
| 8 | GPT-5.4 Nano (2026-03-17) (High) | 29.38 | #266 (GPT-5.4 Nano (2026-03-17)) |
| 9 | Gemini 2.5 Pro | 28.98 | #152 |
| 10 | GPT-5.2 | 28.93 | #110 |
| 11 | Qwen 3 32B (Non-reasoning) | 28.61 | #443 (Qwen 3 32B) |
| 12 | Qwen 3 30B A3B 2507 Instruct | 28.48 | #488 |
| 13 | Gemini 2.5 Flash (Non-reasoning) | 28.48 | #245 (Gemini 2.5 Flash) |
| 14 | Gemini 3.1 Flash Lite (Preview) | 28.43 | #194 |
| 15 | GPT-OSS-20B (Medium) | 27.91 | #518 (GPT-OSS-20B) |
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-mms-sq · How It Works · Data refreshed daily, snapshot 2026-10-20.