3 ms·For small models, tokenization can reach 1-10% of total inference time.by rockinghigh 2mo agoFor small models, tokenization can reach 1-10% of total inference time.