List your company

Per-character Tokenization on MBPP Sanitized

0.32Tokens/Char (MBPP Sanitized)BrahmicTokenizer-131K
0.315840.343920.3720.40008May 28, 2026

Evaluation results

MethodLinks
BrahmicTokenizer-131KVocab=131K2026.050.32
Tekken/Sarvam-mVocab=131K2026.050.338
Sarvam-30BVocab=262K2026.050.388
Sarvam-1Vocab=68K2026.050.424