EuroEval Slovak NLU - ScaLA SK: leaderboard

The "ScaLA SK" subtask of EuroEval Slovak NLU. Slovak-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. 204 models tracked.

Top models

#ModelScoreOverall rank
1Gemini 3 Pro (Preview)69#66
2Gemini 3.1 Flash Lite (Preview)66.71#194
3Gemini 2.5 Flash (Thinking)65.44#245 (Gemini 2.5 Flash)
4Gemini 2.5 Pro64.97#152
5Gemini 3 Flash (Preview) (Thinking)64.02#82 (Gemini 3 Flash (Preview))
6GPT-563.79#98
7Gemini 3 Flash (Preview) (Non-reasoning)63.09#82 (Gemini 3 Flash (Preview))
8Claude Sonnet 4.661.8#90
9Gemma 4 31B (IT)61.46#176
10GPT-5.4 Mini (2026-03-17) (High)61.45#169 (GPT-5.4 Mini (2026-03-17))
11Claude Sonnet 4.5 (Thinking)60.17#142 (Claude Sonnet 4.5)
12Gemini 2.5 Flash Lite (Thinking)58.55#422 (Gemini 2.5 Flash Lite)
13GPT-5 Mini (2025-08-07)58.36#172
14Qwen 3 32B (Non-reasoning)56.94#443 (Qwen 3 32B)
15Qwen 3.5 35B A3B Base56.67#310

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

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