█

AI/TLDR

OpenAI · 2026-09-29 · major

OpenAI Decisions API — GPT-6 Luna picks from your answers in 150 ms

In short

OpenAI's Decisions API runs GPT-6 Luna to pick one of a developer's predefined answers up to 10x faster than the Responses API. Launched at DevDay 2026, it became generally available on October 22 at $0.10 per 1M input tokens.

OpenAI · 2026-09-29 · 6 sources checked

FourWeekMBA graphic explaining OpenAI's Decisions API and its fixed output space
FourWeekMBA

OpenAI's answer to Jev: an endpoint that returns a choice, not text, about ten times faster than a normal Luna call.

Quick facts

MakerOpenAI
Modelgpt-6-luna (only model so far)
LatencyAbout 150 ms (vs 1.6 s for a standard Luna call)
InputText, JSON or images + a closed list of answers
Questions per request1 to 400
Price$0.10 / 1M input tokens; $0.01 cached; no output charge
AvailabilityGenerally available since October 22, 2026

What is it?

The Decisions API is a new OpenAI endpoint, announced at DevDay on September 29, 2026, that does not write free text. The developer sets a question and a fixed list of possible answers, and the API returns which answer fits best. The New Stack and The Decoder frame it as OpenAI's response to Jev, the typed decision model TypeSafe AI released in mid-September.

How does it work?

A specialized version of GPT-6 Luna reads the context — text or an image — and scores only the allowed options, then returns the most likely one with a confidence score. Because the output space is closed, the model cannot answer outside the list. Skipping text generation cuts the response time to about 150 milliseconds, against 1.6 seconds for a standard GPT-6 Luna call.

Why does it matter?

Classification, routing and "what should the agent do next" steps are often the slowest and least reliable part of an LLM app, because the model writes text that then has to be parsed. A fast endpoint that can only return a valid option makes these steps usable in real time. Billing only input tokens also makes high-volume routing cheap, though OpenAI has published no accuracy figure yet.

Who is it for?

developers building classifiers, routers and agent control loops

Frequently asked questions

How much does the OpenAI Decisions API cost?
At general availability, the OpenAI Decisions API with gpt-6-luna costs $0.10 per 1M input tokens and $0.01 per 1M cached input tokens, according to Unite.AI's report on the GA release. OpenAI charges only for input tokens, because the API returns a choice instead of generated text. Regional processing premiums and long-context multipliers still apply.
Is the OpenAI Decisions API generally available?
Yes. The OpenAI Decisions API became available to all customers on October 22, 2026, TechCrunch reports, after launching at DevDay on September 29 and a public beta. Each request now takes up to 400 questions, OpenAI added inference capacity in more regions, and Zero Data Retention and HIPAA use are available to eligible customers. gpt-6-luna is still the only model.
How does the OpenAI Decisions API compare to Jev?
Both the OpenAI Decisions API and TypeSafe AI's Jev answer a question by choosing from predefined options instead of generating text. The New Stack describes the Decisions API as OpenAI's answer to Jev. OpenAI's version runs as a hosted API on a specialized GPT-6 Luna and has no published accuracy figure yet, so a direct comparison needs your own tests.
What can the OpenAI Decisions API be used for?
OpenAI says developers supply context as text or images and get back answers they can use to classify content, route requests, or choose an agent's next action. A typical case is a support app deciding whether a message goes to technical support, sales, or complaints — the Decisions API returns the category directly, with a confidence score.

Sources · 6 outlets

Tags

  • openai
  • decisions-api
  • gpt-6-luna
  • classification
  • routing
  • decision-models
  • jev
  • devday
  • api
  • launch

← All releases · Learn AI