OpenAI makes its Decisions API available to all customers

In short: OpenAI says its Decisions API is now available to all customers, offering a faster, cheaper way for apps to make simple choices.

What happened

OpenAI has made its Decisions API model widely available for the first time. This product is meant for tasks where software needs to choose between options quickly, like picking the best answer from a short list and saying how confident it is.

TechCrunch reports that this moves OpenAI into a growing market sparked by TypeSafe’s Jev model. These tools are sometimes described as “decision models” or “System One models,” meaning they focus on quick decisions instead of long, chat-style replies (think of it like a cashier who can instantly total your items, rather than a clerk who writes an essay about prices).

According to OpenAI, the key change is speed. OpenAI product lead Nikunj Handa told TechCrunch the company reduced response time from 1 to 3 seconds down to roughly 100 to 180 milliseconds (milliseconds are tiny fractions of a second). OpenAI did this by reworking its ChatGPT Luna model to run faster and cost less, rather than building a brand-new model from scratch.

OpenAI announced the Decisions API at its developer event on September 29, and it became available to all customers on Thursday. The public version also adds support for voice and images, and OpenAI says it has worked on reducing delays in different regions.

Why it matters

Many everyday apps rely on quick background decisions, like sorting messages, spotting suspicious activity, or deciding what to show next. If these decisions happen faster and cost less, companies may add more of them, which can make apps feel more responsive and more automated.

Source: TechCrunch AI