OpenAI fully launches Decisions API with up to 10x faster responses

OpenAI has fully launched its Decisions API. Built on GPT-6 Luna, the API focuses on questions with bounded answers, helping apps and agents make near real-time decisions from text, images, and structured inputs. Responses are up to 10 times faster than calling the model through the Responses API. On pricing, the API costs $0.1 per million input tokens, and just $0.01 per million for cached inputs. It only charges for input tokens, while output tokens are free. Teams across OpenAI are already using it in products and workflows. Since the public beta, the API has added native GPT-Live integration, prompt caching for reused context, and inference data residency in the US and EU. Developers can now send native JSON inputs, include images in instructions, and submit up to 400 questions per request, double the previous limit. OpenAI has also expanded capacity across Asia to keep cutting delays.