v1.2.0
Fast and lightweight token estimation for any LLM without requiring a full tokenizer
Source code on GitHub