OpenAI fully opens the decision-making API, input one million tokens for 0.1 USD

On October 22 local time, OpenAI announced the full launch of the Decision API. This API is based on GPT-6 Luna and is designed to assist applications and agents in making near real-time decisions based on text, images, and structured inputs for questions with limited answers. IT Home reported that its response speed is up to 10 times faster than calling GPT-6 Luna through the Responses API.

In terms of pricing, the cost is $0.1 per million tokens for input, which is approximately 0.67 Chinese Yuan at the current exchange rate; for cache hit input, it is $0.01 per million tokens, about 0.067 Chinese Yuan. The API charges only for input tokens, while output tokens are free. The page states that OpenAI has already used this API internally for product and business workflows.

New features after the public beta include native GPT-Live integration, prompt caching for reused contexts, and reasoning data residency in the United States and the European Union. Developers can pass in native JSON and include images in the question instructions, with a maximum of 400 questions submitted in a single request, double the previous limit. The page also states that OpenAI is continuously optimizing latency and expanding capacity in several regions across Asia.