EuroEval Slovak NLU - UNER SK: leaderboard

The "UNER 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: Micro F1 excluding MISC (%). Source: euroeval.com. Saturation forecast: Around January 2027. 204 models tracked.

Top models

#ModelScoreOverall rank
1Qwen 3 4B 2507 (Thinking)73.36#538 (Qwen 3 4B 2507)
2Qwen 3 235B A22B 2507 FP8 (Thinking)73.22#246 (Qwen 3 235B A22B 2507 FP8)
3Gemma 4 31B (IT)72.84#176
4Voxtral-Small-24B-250768.31#581
5GPT-5.4 Mini (2026-03-17) (High)67.46#169 (GPT-5.4 Mini (2026-03-17))
6Llama 3.1 70B66.78#592
7Mistral Small 3.265.96#605
8Mistral Small 3.165.77#610
9GPT-OSS-20B (Medium)65.77#518 (GPT-OSS-20B)
10Gemma 4 26B A4B (IT)65.53#297
11Claude Sonnet 4.5 (Non-reasoning)64.49#142 (Claude Sonnet 4.5)
12Gemini 3 Flash (Preview) (Non-reasoning)64.41#82 (Gemini 3 Flash (Preview))
13Gemini 2.5 Flash (Non-reasoning)64.02#245 (Gemini 2.5 Flash)
14Claude Sonnet 4.5 (Thinking)63.92#142 (Claude Sonnet 4.5)
15GPT-563.83#98

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-uner-sk · How It Works · Data refreshed daily, snapshot 2026-10-20.