Token Count Estimator

Token Count Estimator

Large language models process text as tokens, not words - this calculator gives a quick rule-of-thumb estimate for planning purposes.

Estimated Tokens = Word Count x 1.3

Example

A 1,000-word document:

Estimated Tokens = 1,000 x 1.3 = ~1,300 tokens

Why the 1.3 Multiplier

A commonly cited rough approximation for English text is roughly 1.3 tokens per word - equivalently, about 4 characters per token on average across typical English prose, based on how modern subword tokenizers tend to split common words into single tokens while splitting rarer or longer words into multiple tokens.

Important Caveats

• Tokenizer-specific: exact ratios vary by the specific tokenizer used by different model providers.
• Language-dependent: non-English languages, especially those using different scripts, often tokenize considerably less efficiently than English.
• Content-dependent: technical jargon, rare words, and code tend to use more tokens per word than common everyday language.

Use this as a quick planning estimate, not an exact count - for precise token counts, use the specific tokenizer library corresponding to the model you're actually working with.