OpenAI made its Decisions API available to all users with API access on October 22, 2026, pricing it at $0.10 per million input tokens. Outputs, cache reads and cache writes are free. Built on the GPT-6 Luna model, the tool provides structured responses in the form of probabilities, choices and scores for routing, classification and repeated micro-decisions at scale. Each request can contain up to 400 questions, compared with a lower capacity during the beta, and the service achieved a latency of approximately 150 milliseconds in demonstrations. OpenAI also reports speeds up to ten times faster than the standard Responses API for bounded decisions, alongside support for images, JSON, prompt caching and Zero Data Retention.
Source: Read the original article
