monolingual encoder-only language models
AI & ML interests
Web as a corpus, Large Language Models, Machine Translation, Language Technologies, Natural Language Processing
Recent Activity
Organization Card
Our project name, HPLT, is an acronym for High Performance Language Technologies. We are aiming high at combining large quantities of data, a number of languages and high-performance computing to build powerful and efficient language and translation models. Another goal of HPLT is to publish the results of this project in a shared space with open licenses.
Collections
7
Translation models trained on OPUS data including HPLT datasets
-
Helsinki-NLP/opus-mt-tc-bible-big-aav-fra_ita_por_spa
Translation • Updated • 130 • 2 -
Helsinki-NLP/opus-mt-tc-bible-big-afa-en
Translation • Updated • 175 • 1 -
Helsinki-NLP/opus-mt-tc-bible-big-afa-deu_eng_fra_por_spa
Translation • Updated • 175 -
Helsinki-NLP/opus-mt-tc-bible-big-afa-deu_eng_nld
Translation • Updated • 125
models
262
HPLT/hplt_bert_base_bhr-Latn
Updated
HPLT/hplt_bert_base_cat-Latn
Updated
HPLT/hplt_bert_base_slk-Latn
Updated
HPLT/hplt_bert_base_rus-Cyrl
Updated
HPLT/hplt_bert_base_ron-Latn
Updated
HPLT/hplt_bert_base_por-Latn
Updated
HPLT/hplt_bert_base_pol-Latn
Updated
HPLT/hplt_bert_base_pes-Arab
Updated
HPLT/hplt_bert_base_nob-Latn
Updated
HPLT/hplt_bert_base_nno-Latn
Updated