On October 22, 2026, OpenAI said the Decisions API is generally available. It was previously in public beta. The API is built with GPT-6 Luna and is aimed at questions with bounded answers, such as classification, routing, and scoring, from text, images, and structured inputs. DigIntoAI has not tested the Decisions API.
OpenAI writes that the Decisions API helps applications and agents make near real-time decisions up to 10× faster than GPT-6 Luna through the Responses API. That speed comparison is OpenAI's claim. For DigIntoAI notes on Luna and related OpenAI pricing, see GPT-6 Sol vs Luna and GPT-6.1 Sol Ultrafast pricing.
Pricing (OpenAI)
Per OpenAI, Decisions API pricing is $0.10 per million input tokens and $0.01 per million cached input tokens. There are no output-token charges. OpenAI points readers to its pricing guide for cached-write detail. DigIntoAI has not verified billing on a live account.
What changed since public beta
Since public beta, OpenAI says it added a native GPT-Live integration, prompt caching for reused context, and inference residency in the US and EU. Developers can also pass native JSON inputs and include images in question instructions. The maximum number of questions per request rose from 200 to 400. For prior DigIntoAI coverage of OpenAI prompt caching, see OpenAI 12-hour prompt caching. For Luna on ChatGPT Free and Go, see GPT-6 Free and Go Luna.
We have not tested the Decisions API, its latency claims, or the listed post-beta features.
