Token estimator
Paste any text and get a quick estimate of how many tokens it uses, and roughly how a model would split it.
Characters148
Words26
≈ Tokens37
Approximate split · first 160 tokens
You are a support assistant. Read the ticket below, summarise it in two sentences, then classify its urgency as low, medium or high. Answer in JSON.
Ratio used: about 4 characters per token. A quick estimate: exact counts depend on each model’s tokenizer.
How it works
Models don’t read words, they read tokens: chunks of a few characters. Knowing the count helps you stay within limits.
[Explain the method in two short paragraphs: the ratio used per text type, why French and code differ from English, and when to use an exact tokenizer instead.]
v[1.0]First release[Date]