OpenAI launches Decisions API public beta on GPT-6 Luna
TL;DR
- OpenAI opened a dedicated Decisions API in public beta on October 6, 2026, running only on gpt-6-luna with text and image inputs.
- The endpoint is priced at $0.10 per 1M input tokens, with cache reads, cache writes, and output tokens all unbilled.
- It returns three typed shapes — predicate (0–1), choice with confidence, and score across ordered levels — and ships with HIPAA eligibility and US/EEA residency.
OpenAI opened a dedicated Decisions API in public beta on October 6, running only on gpt-6-luna. The announcement on the OpenAI developer forum prices it at $0.10 per 1M input tokens and charges nothing for output tokens or for cache reads and writes.
The endpoint hits POST /v1/decisions and returns typed answers in three shapes: a predicate as a 0-to-1 probability, a choice from fixed options with confidence scoring, and a score rated against ordered severity levels. Text and images are both accepted, but as Unite.ai notes, image inputs must be inline base64 data URLs — hosted image links and uploaded file IDs are refused.
The pitch is speed. OpenAI says the endpoint "produces those answers about 10x faster than the Responses API." General availability is expected "in the coming weeks."
Alongside the price-per-token line, the beta ships with Zero Data Retention support, HIPAA eligibility, and data residency in the US and Europe (EEA/Switzerland). Those compliance checkboxes make it a plausible drop-in for moderation and gating flows that previously stitched together JSON-schema prompts on Chat Completions.
The 10x figure is measured against OpenAI's own Responses API, not an external vendor. Neither the forum post nor the follow-on coverage publishes a head-to-head accuracy number comparing /v1/decisions to the same gpt-6-luna model called through the Responses API.
Originally reported by community.openai.com
Read the original article →Original headline: OpenAI Opens Decisions API Public Beta on GPT-6 Luna, Prices Input at $0.10 per Million Tokens