Loading
Models bill by tokens. As a rough guide, one token is about 4 characters of English. Write `estimate_tokens(text)` that returns the number of characters divided by 4, rounded up. Empty text is 0 tokens.
len(text) gives characters. math.ceil rounds up.