ToolingPublished All news from that day
Ongoing case: OpenAI's models Built on a model · 21 stories
OpenAI made its Decisions API generally available, with US and EU inference residency.
The API, built on GPT-6 Luna, returns a probability, a choice or a score for a bounded question. OpenAI says it is up to 10x faster than GPT-6 Luna through the Responses API.
Check our sources · 9 facts from 1 sourceOpenAI said on 22 October 2026 that the Decisions API is generally available. It is built with GPT-6 Luna and is meant for questions with bounded answers, from text, images and structured inputs.
Developers define a question and its allowed answers, and the API returns one of three answer types: a probability for a predicate, a choice among supplied values, or a score against a rubric.
OpenAI says it is up to 10x faster than GPT-6 Luna through the Responses API. Since the public beta OpenAI added a native GPT-Live integration, prompt caching for reused context, and inference residency in the US and the EU. The limit is now 400 questions per request, twice the earlier limit.
The speed and cost figures that customers such as Sierra and Ramp give on the page are their own statements, quoted by OpenAI.
Your reaction
We count reactions per story and day, never who reacted. The counts help us choose what goes in the monthly issue. If you are signed in, your own page shows yours too.
We count reactions per story and day, never who reacted. They help choose what goes in the monthly issue.
Check our sources
Every sentence above is checked against this source.
1 Build more responsive apps with the Decisions API
Open the source archived copy-
OpenAI's post of 22 October 2026 says it is making the Decisions API generally available.
citeOctober 22, 2026 ... Today, we’re making the Decisions API generally available.
-
OpenAI says the Decisions API is built with GPT-6 Luna and focuses Luna on questions with bounded answers, using text, images and structured inputs.
citeBuilt with GPT‑6 Luna, the Decisions API focuses Luna’s intelligence on questions with bounded answers, helping applications and agents make near real-time decisions from text, images, and structured inputs
-
OpenAI says the Decisions API makes near real-time decisions up to 10x faster than GPT-6 Luna through the Responses API.
citehelping applications and agents make near real-time decisions from text, images, and structured inputs up to 10x faster than GPT‑6 Luna through the Responses API.
-
OpenAI says that since the public beta it has added a native GPT-Live integration, prompt caching for reused context, and inference residency in the US and EU.
citeSince public beta, we’ve added a native GPT‑Live integration, prompt caching for reused context, and inference residency in the US and EU.
-
OpenAI says developers can pass native JSON inputs, include images in question instructions, and ask up to 400 questions per request, twice the previous limit.
citeDevelopers can also pass native JSON inputs, include images in question instructions, and ask up to 400 questions per request, twice the previous limit.
-
OpenAI says developers define a question and its allowed answers, then give context as text, images or native JSON.
citedevelopers define a question and its allowed answers, then provide context as text, images, or native JSON.
-
OpenAI lists three answer types: predicate, which returns a probability from 0 to 1; choice, which returns one of the supplied values with probabilities and confidence; and score, the probability-weighted average of rubric level indices.
citepredicate ... probability: an estimate from 0 to 1 that the condition is true. choice ... choice: one of your supplied values. The response also includes probabilities and confidence. score ... score: the probability-weighted average of the level indices.
-
In a quote on OpenAI's page, Ajeet Grewal of Sierra says Sierra runs the Decisions API at under 280 ms p95, about 3x faster and 90% lower cost per call than its previous approach. the company says
citeIt runs at under 280 ms p95, about 3× faster and 90% lower cost per call than our previous approach, with no drop in customer outcomes.”—Ajeet Grewal, Engineer, Sierra
-
In a quote on OpenAI's page, René Sultan of Ramp says the Decisions API made attributing AI costs 4x faster and 30% cheaper than GPT-6 Luna, without losing accuracy. the company says
citeThe Decisions API has made it 4x faster and 30% cheaper to attribute our AI costs to the actual work when compared to GPT‑6 Luna, without sacrificing accuracy.”—René Sultan, Research Engineer, Ramp
Every sentence above is checked against this source. Each fact has its own address you can cite.
Source 1 · OpenAI · 22 Oct 2026
Build more responsive apps with the Decisions API
OpenAI's post of 22 October 2026 says it is making the Decisions API generally available.
October 22, 2026 ... Today, we’re making the Decisions API generally available.
OpenAI says the Decisions API is built with GPT-6 Luna and focuses Luna on questions with bounded answers, using text, images and structured inputs.
Built with GPT‑6 Luna, the Decisions API focuses Luna’s intelligence on questions with bounded answers, helping applications and agents make near real-time decisions from text, images, and structured inputs
OpenAI says the Decisions API makes near real-time decisions up to 10x faster than GPT-6 Luna through the Responses API.
helping applications and agents make near real-time decisions from text, images, and structured inputs up to 10x faster than GPT‑6 Luna through the Responses API.
Show all 9 factsShow fewer facts
OpenAI says that since the public beta it has added a native GPT-Live integration, prompt caching for reused context, and inference residency in the US and EU.
Since public beta, we’ve added a native GPT‑Live integration, prompt caching for reused context, and inference residency in the US and EU.
OpenAI says developers can pass native JSON inputs, include images in question instructions, and ask up to 400 questions per request, twice the previous limit.
Developers can also pass native JSON inputs, include images in question instructions, and ask up to 400 questions per request, twice the previous limit.
OpenAI says developers define a question and its allowed answers, then give context as text, images or native JSON.
developers define a question and its allowed answers, then provide context as text, images, or native JSON.
OpenAI lists three answer types: predicate, which returns a probability from 0 to 1; choice, which returns one of the supplied values with probabilities and confidence; and score, the probability-weighted average of rubric level indices.
predicate ... probability: an estimate from 0 to 1 that the condition is true. choice ... choice: one of your supplied values. The response also includes probabilities and confidence. score ... score: the probability-weighted average of the level indices.
In a quote on OpenAI's page, Ajeet Grewal of Sierra says Sierra runs the Decisions API at under 280 ms p95, about 3x faster and 90% lower cost per call than its previous approach. the company says
It runs at under 280 ms p95, about 3× faster and 90% lower cost per call than our previous approach, with no drop in customer outcomes.”—Ajeet Grewal, Engineer, Sierra
In a quote on OpenAI's page, René Sultan of Ramp says the Decisions API made attributing AI costs 4x faster and 30% cheaper than GPT-6 Luna, without losing accuracy. the company says
The Decisions API has made it 4x faster and 30% cheaper to attribute our AI costs to the actual work when compared to GPT‑6 Luna, without sacrificing accuracy.”—René Sultan, Research Engineer, Ramp
Topics
The morning email
On the mornings we publish: the three top stories and up to four short ones. Free.
We email you a link to confirm. An issue may include one sponsor, always labelled Sponsored · Advertisement. Our emails count opens and clicks, not who made them. Unsubscribe in one click. What we keep