HPLT/HPLT3.0
This is a large-scale collection of web-crawled documents in 198 world languages, produced by the HPLT project. The source of the data is Internet Archive and Common Crawl. For a detailed description of this and previous releases by HPLT, please refer to our website. NB: the HPLT datasets are not hosted on HuggingFace! See download instructions below. HPLT release v3.0 In July 2025, the European HPLT initiative has completed a new release of its monolingual datasets, offering… See the full description on the dataset page: https://huggingface.co/datasets/HPLT/HPLT3.0.
| Author / organization | HPLT |
|---|---|
| Downloads | 277 (as reported by Hugging Face) |
| Likes | 24 |
| Created | 27 October 2025 |
| Last updated at source | 8 July 2026 |
Popularity, last 2 days
277 now (+8 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.
Connected in the knowledge graph
- HPLT/NorOLMo-13BModel · used to train
- HPLT/FinOLMo-13BModel · used to train
- permitt/galton-modernbertic-largeModel · used to train
More from HPLT
- HPLT/HPLT2.0_cleanedDataset
- HPLT/DocHPLTDataset
- HPLT/DocHPLT-v3-cleanDataset
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.