OpenAI Decisions Playground Lets Developers Test Three Question Types

OpenAI’s Decisions Playground lets developers test how its public-beta API handles text and images before they write code. It offers three types of questions that help sort information into groups, send requests to the right place, and decide which tasks are most important.

A predicate question estimates how likely it is that a condition is true. A choice question picks one option from a list provided by the developer. A score question rates the input using rules the developer has defined. Developers can try different questions and inputs in the Playground before adding them to an application.

The Decisions API currently works only with gpt-6-luna. Input starts at $0.10 per million tokens, and there is no extra charge for output tokens in Decisions requests. However, processing in certain regions and higher prices for long inputs can increase the input cost.

OpenAI released the Decisions API in public beta on Oct. 6.