OpenAI makes the GPT‑6 Luna-powered Decisions API generally available

What is changing?

OpenAI made the Decisions API generally available on October 22. Powered by GPT‑6 Luna, the service focuses on bounded decisions such as classifying an input, routing a request or selecting the next action from text, images and structured data.

Developers define the question, allowed answers and relevant context. The API can return a probability, choose from predefined options or produce an ordered score. OpenAI says it can be up to 10 times faster on certain workloads than using GPT‑6 Luna through the Responses API.

The general release adds native GPT‑Live integration, prompt caching for reused context, inference residency in the US and EU, native JSON inputs and support for up to 400 questions in one request. Input costs $0.10 per million tokens, with no output-token charge.

The speed and accuracy examples come from OpenAI and early customers. Production teams should measure decision thresholds, low-confidence cases, latency and error costs on their own data, while preserving human review or a safe fallback for critical workflows.