FAQs

Do I own my translations?

Yes.

You, the customer, retain full ownership and rights to any translated output produced through the Services. We make no claim to your output.

For more, see our Terms of Use.

What are tokens?

When text is sent to GT APIs, it is broken down into chunks called tokens. Think of tokens as small pieces of your input text, like words or parts of words, that an AI model uses to understand and translate your content.

A helpful rule of thumb is that one token generally corresponds to ~4 characters of text for common English text. This translates to roughly ¾ of a word (so 100 tokens ~= 75 words).

For example, the string Hello, world! has 13 characters, and 4 tokens: Hello, ,, world, and !.

If you go over the token limit for your plan, we begin to charge based on token usage. We only ever charge based on the length of your input — never the length of the translation. Read more about our available plans here.

For example, imagine you have the string "Hello, world!", and your project's languages are "fr" (French), "es" (Spanish), and "de" (German). Token usage will be calculated as 4 * 3 = 12. That is, the 4 tokens of "Hello, world!", translated 3 times.

We currently tokenize with OpenAI's tiktoken, which is the same tokenizer used by GPT-4. You can explore this tokenizer here.