Recommended language models (larger and higher-quality than Hindi-Bert)
#1
by
monsoon-nlp
- opened
Hindi-BERT is a smaller Electra model which I uploaded in summer 2020. Recommended alternatives as of May 2022:
For masked language models, I recommend Google's MuRIL model trained on English, Hindi, and other major Indian languages, both in their script and latinized script. Two available sizes:
For causal language models, I recommend SberBank / mGPT, though this is a large model