5/26/2023 0 Comments Apopen prompts![]() ![]() You can find details on token pricing here. Requests to these different engines are priced differently. Each engine has a spectrum of capabilities, with davinci being the most capable and ada the fastest. The API offers multiple engine types at different price points. The following limits on tokens are useful to keep in mind when using the API:Ĭompletions - depending on the engine used, requests can use up to 4000 tokens shared between prompt and completion.įor specialized endpoints - Answers, Search, and Classifications - the query and longest document must be below 2000 tokens together. To further explore how text is broken into tokens, you can experiment with our interactive Tokenizer tool. The higher token-to-char ratio can make it more expensive to implement the API for languages other than English. For example ‘Cómo estás’ (‘ How are you’ in Spanish) contains 5 tokens (for 10 chars). How words are split into tokens is also language-dependent. The transcript of the US Declaration of Independence contains 1,695 tokens. Wayne Gretzky’s quote " You miss 100% of the shots you don't take" contains 11 tokens. To get additional context on how tokens stack up, consider this: Here are some helpful rules of thumb for understanding tokens in terms of lengths: These tokens are not cut up exactly where the words start or end - tokens can include trailing spaces and even sub-words. ![]() Before the API processes the prompts, the input is broken down into tokens. Tokens can be thought of as pieces of words. ![]()
0 Comments
Leave a Reply. |