OpenAI makes the Decisions API generally available with GPT-6 Luna

OpenAI has made the Decisions API generally available. Built on GPT-6 Luna, it targets fast, bounded-answer decisions such as classification, routing, and next-action selection from text, images, and JSON. The API returns a probability, choice, or score. Updates since beta include native GPT-Live integration, prompt caching, US and EU inference residency, native JSON input, and a 400-question-per-request limit. Pricing is $0.10 per million input tokens, with cached input at $0.01 per million.

Key takeaways

• Decisions API is generally available and built with GPT-6 Luna, focused on questions with bounded answers.
• OpenAI says it runs up to 10x faster than GPT-6 Luna through the Responses API.
• Supports three answer types: probability, choice, and score, with up to 400 questions per request.
• Costs $0.10 per million input tokens and $0.01 per million cached input tokens; no output-token charges.

Why it matters

Lower-latency, lower-cost routing and classification decisions could make small per-step model calls practical in real-time workflows such as voice agents and agent orchestration.

Read with care

Speed and cost figures come from OpenAI and customer testimonials in the announcement; source claims not independently checked.