What the Decisions API is
The Decisions API is OpenAI’s hosted API for typed decisions. It does the job of a System One model, though OpenAI does not use that term. OpenAI announced it at DevDay on 29 September 2026. You define questions, each with a fixed list of possible answers, and send context as text or images. The API returns an answer from your list. OpenAI says it runs on GPT-6 Luna, the cheapest model in the GPT-6 family.
What it returns
OpenAI’s posts say only that you get an answer back, “a selection” in its X thread. The New Stack reports predefined answers with confidence scores. OpenAI has not published the response format, whether the scores are probabilities, whether they are calibrated, or whether the API supports Score or Noul questions. Only a Choice is confirmed, so this page lists only that.
What it is good at
OpenAI names three jobs: classify content, route requests, and choose an agent’s next action. Its X thread gives a support request and the teams it could go to as the example. Those map to support inbox triage, intent and model routing and agent routing and skill selection.
How it compares with Jev
Both are hosted APIs that pick from answers you define. The Decisions API accepts images as well as text, which no other model on this site claims. Jev reads text, publishes a price and a request format, and returns calibrated probabilities.
Every, which had preview access, ran early tests. On a text replay of computer tasks, the Decisions API picked the right control in 76 of 78 steps against 73 for Jev. On a thread-classification test the two tied on accuracy and Jev was faster. Every says the Decisions API beat Jev on a few early tests but trailed on its broader evals. These are the testers’ own results.
Two questions stay open until OpenAI publishes a reference. Does it return a probability for every option, as Jev does? And does it accept the /v1/systemone request shape that Jev and other models share?
What is not known yet
As of 30 September 2026, OpenAI has published no docs page, endpoint, price, limits, SDK support, benchmark or statement on calibration. The API does not appear in OpenAI’s docs changelog, in the latest Python and Node SDK releases, on OpenRouter or in the Hugging Face Decision Index. The New Stack also lists whether developers can tune it on their own data as unclear. An OpenAI spokesperson told The New Stack that OpenAI plans to share more at broad rollout.
Access
Preview access is limited to selected API customers. OpenAI says broad release is planned in the coming days, while Every expects a launch in a few weeks. This page will be updated when OpenAI publishes docs and pricing.
Specifications
| Question types | Choice |
| Max Choice options | Not documented |
| Score levels | Not documented |
| Questions per call | Not documented |
| Total context | Not documented |
| State budget | Not documented |
| Rate limit | Not documented |
| SDKs |
OpenAI has published no limits for the Decisions API. The New Stack lists the number of answers per request as unclear. GPT-6 Luna itself has a 1,050,000-token context window, but OpenAI has not said the Decisions API offers the same window. Checked on 2026-09-30.
Versions
- Limited preview, 29 Sep 2026, OpenAI names no version or model id. It describes the API as built on GPT-6 Luna, and The Decoder describes a version of Luna made for it. Preview access is limited to selected API customers. Release notes
Use cases
What people use OpenAI Decisions API for, one page per pattern.
Support inbox triage with System One models
Send a support ticket to Jev once with every question attached. Category comes back as a selected label, severity and frustration as numbers on scales you wrote, refund intent as a probability. Your code reads those values and decides what happens to the ticket.
Intent and model routing with System One models
One Jev call reads an incoming request and returns its intent as a label plus a difficulty rating on a scale you wrote. Your router reads both numbers and picks the handler: deterministic code, a cheap model, an expensive one, or a human queue.
Agent routing and skill selection with System One models
An agent choosing from a long skill roster reads one truncated line per entry and often loads the wrong thing. Jev ranks every entry in one request and separately answers whether any skill applies at all, so the agent gets a short hint instead of a guess.