MTEB logo MTEB
Slovak icon

Slovak

Cite this benchmark

@misc{suppa2026skmteb,
  archiveprefix = {arXiv},
  author = {Marek Šuppa and Andrej Ridzik and Daniel Hládek and Natália Kňažeková and Viktória Ondrejová},
  eprint = {2606.13647},
  primaryclass = {cs.CL},
  title = {SkMTEB: Slovak Massive Text Embedding Benchmark and Model Adaptation},
  url = {https://arxiv.org/abs/2606.13647},
  year = {2026},
}

Slovak Massive Text Embedding Benchmark (SkMTEB), the first comprehensive benchmark for text embeddings in Slovak, a low-resource West Slavic language. SkMTEB consists of 31 tasks across 7 task types, with tasks adapted from both existing and novel resources.

Reference paper →
Languages 3
Tasks 31
Task Types 7
Models 0