Horizon-Labs/multilingual-zeroshot-synthetic

Multilingual Zero-Shot Classification (synthetic) 253,346 texts in 33 languages, each with a set of candidate labels, the correct label(s), and often a hypothesis template. Built to train multilingual zero-shot (NLI-style) classifiers; also usable for few-shot or instruction data, label-set generalization studies, and multilingual intent/topic/sentiment classification. This is the classification part of the training data of multilingual-zeroshot-small / base / large. Code:… See the full description on the dataset page: https://huggingface.co/datasets/Horizon-Labs/multilingual-zeroshot-syntheti

<iframe src="https://silytics.si/embed/item/45b48bc7-e1a9-47b6-b865-8867f0e567ff" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 19 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.