What a token is
AI models don't read letters or words. They read tokens, which are common pieces of words. In English a token is about four characters, or three quarters of a word, but code, numbers and other languages split differently. Models charge and limit by tokens, so knowing the count tells you whether a prompt fits and what it costs.
Exact counts for OpenAI models
This counter uses the same tokenizers as OpenAI's models: o200k for GPT-4o, GPT-4.1, GPT-5 and the o-series, and cl100k for GPT-4 and GPT-3.5. Chat messages add a few tokens of formatting per message on top of these counts.
Context windows
The bars show how much of common context window sizes your text uses. Leave room for the model's answer: a prompt that fills the whole window leaves no space to reply.
Privacy
The tokenizer is downloaded to your browser and runs here. Your text is never uploaded.