AI4BharatIndicBERT
IndicBERT v2 Samanantar TLM
This 278M-parameter IndicBERT v2 release uses a distinct training configuration: Masked-language training plus translation-language modeling with the Samanantar parallel corpus.
The objective distinguishes this checkpoint
The model card links separate weights for the MLM, Samanantar, back-translation and shared-script experiments. These are training variants, not different service tiers.IndicBERT v2 Samanantar TLM official model card
Control the downstream comparison
Practical guidance. This variant helps study whether parallel sentences improve cross-language transfer. Compare it against MLM-only on the same language pair.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | Downloadable |
| Model identifiers | IndicBERTv2-MLM-Sam-TLM |
| License | MIT.IndicBERT v2 Samanantar TLM official model card |
| Evaluation | IndicXTREME; the release supports downstream classification, entity extraction, question answering and retrieval research.IndicBERT v2 Samanantar TLM official model card |
Related models
- IndicBERT v2 Masked Language Model
AI4Bharat ยท IndicBERT
Multilingual IndicBERT v2 encoder trained with the masked-language-model objective.
Launch date not confirmedDownloadable