List your company

Tokenization on AI4Bharat Sangraha Bengali

1,638Token Count (M)BrahmicTokenizer-131K
1,267.761,363.881,4601,556.12May 28, 2026

Evaluation results

MethodLinks
BrahmicTokenizer-131KEvaluation Protocol=27M-document Indic pretraining corpus2026.051,638—
Sarvam-1Evaluation Protocol=27M-document Indic pretraining corpus2026.051,524-6.9
Sarvam-30BEvaluation Protocol=27M-document Indic pretraining corpus2026.051,282-21.7
Tekken/SmEvaluation Protocol=27M-document Indic pretraining corpus2026.05—20.5