Aranese MT
Translation into and out of Aranese (Occitan of the Val d'Aran) with MADLAD-400: Catalan, Spanish, English, French.
Translation • 3B • Updated • 16Note Start here. MADLAD-400-3B-MT fine-tuned to translate into Aranese with the target tag <2oc_aranes> (from Catalan, Spanish, English and French) and out of Aranese with <2ca>, <2es>, <2en>, <2fr>. Use transformers 4.x.
Ugiat/madlad400-3b-mt-oc_aranes-ct2
Translation • Updated • 12Note The same model converted to CTranslate2 int8 (2.9 GB): faster and lighter, for production translation on GPU or CPU.
google/madlad400-3b-mt
Translation • 3B • Updated • 148k • 245Note Upstream: the original multilingual translation model this work fine-tunes (Google).
BSC-LT/Catalan-Aranese_Parallel_Corpus
Viewer • Updated • 539k • 71 • 2Note Upstream: Catalan-Aranese parallel corpus (BSC).
projecte-aina/ES-OC_Parallel_Corpus
Viewer • Updated • 420k • 53 • 2Note Upstream: Spanish-Aranese parallel corpus (Projecte Aina).
MADLAD-400: A Multilingual And Document-Level Large Audited Dataset
Paper • 2309.04662 • Published • 26Note Upstream: the MADLAD-400 paper (Kudugunta et al., 2023), which introduced the base model.