AI Resources
AfroXLMR vs Zabantu-XLM-Roberta
A verified, side-by-side comparison. Both records are status-checked by Findra, so you are comparing what each actually offers today, not a stale listing.
Category
AI Resources
AI Resources
Type
NLP Model
NLP Model
Country
🌍 Pan-African
🇿🇦 South Africa
Docs status
Docs live
Docs live
Licensing
Pricing
Not listed
Open weights
Verified
Verified
Unverified
Last verified
5 Jul 2026
5 Jul 2026
Tags
nlp, multilingual, african-languages, masked-language-model, xlm-roberta
south-africa, xlm-roberta, bantu-languages, tshivenda, zulu
Summary
AfroXLMR is an XLM-R-large model (0.6B params) adapted to African languages via multilingual adaptive fine-tuning, covering 17 African languages plus Arabic, French, and English. Created by David Adelani (Davlan) and published at COLING 2022 for cross-lingual transfer tasks like NER.
Zabantu is a family of XLM-RoBERTa masked language models (roughly 80M to 250M params) trained from scratch on South African Bantu languages including Tshivenda, Zulu, Xhosa, Swati, Northern and Southern Sotho, Setswana and Xitsonga. It serves as a benchmark for low-resource Bantu language NLP. It was built by the Data Science for Social Impact group at the University of Pretoria.