Anthropic Claude Tokenizer

Count tokens for Claude models (Opus, Sonnet, Haiku) using the official tokenizer

Input Text

⋯

Enter some text above to see the token count

Type or paste text in the input field above to analyze tokens and character count.

About Claude's Tokenizer

Claude uses a BPE (Byte Pair Encoding) tokenizer similar to GPT models
Average token length: ~4 characters for English text
Special characters and code may use more tokens