TypeSafe API.
A decision model that answers typed questions about text you supply, with calibrated probabilities. Compare endpoints and pay-as-you-go pricing.
28 providers 1,490 endpoints
Send one piece of state — a document, a chat log, a ticket, any JSON — together with a map of typed questions, and get back one typed answer per question with calibrated probabilities. A choice question picks one option from up to 255 you define and returns the full probability distribution plus a confidence value; a score question rates the state on an ordered rubric of 2 to 10 levels you write; a noul question returns the probability that a statement about the state is true. Every question is evaluated in parallel against the same state in a single call, so asking thirteen questions at once costs far less than thirteen calls. One request takes up to 64,000 tokens, and a larger one is rejected and not billed; send only the state your questions need, because anything beyond it lowers accuracy and adds to the token count. The model returns decisions, not prose: it never writes text, it is not a calculator, and it judges what a passage says rather than reasoning in several steps.