muchad/mdeberta-hybrid-30k
0.1B
•
Updated
•
37
Pruned mDeBERTa
A collection of vocabulary-pruned mDeBERTa-v3 models (30k/20k) optimized for efficient Indonesian NLP deployment.
A collection of vocabulary-pruned mDeBERTa-v3 models (30k/20k) optimized for efficient Indonesian NLP deployment.