Quick reference (English text)
| Text | Words | Approx. tokens |
|---|---|---|
| A short paragraph | 100 | about 130 |
| One page | 500 | about 670 |
| A 10 page document | 5,000 | about 6,700 |
| A short book | 50,000 | about 67,000 |
Rule of thumb for English: 1 token is about 4 characters or three quarters of a word. Code, numbers and many other languages use more tokens for the same length. This is an estimate, not an exact count.
Words, characters and tokens
AI models read text as tokens, not words. In English, one token is roughly four characters or about three quarters of a word. That means 1,000 words is close to 1,300 tokens.
How to use the converter
Choose a direction, enter an amount and pick a unit. You get an approximate result with a likely range. Use the reverse mode to see how much text a token limit allows.
When the estimate is off
Code, numbers, emoji and many languages other than English break into more tokens per word. For exact counts use the tokenizer or the usage data from your AI provider.
FAQ
No. It uses a common rule of thumb for English. Real counts vary by model and by text.
Multiply the token limit by 0.75 for a rough number of English words. Remember that the reply uses tokens too.