OpenAI Makes Decisions API Generally Available
OpenAI has made Decisions API generally available to all customers, adding voice and visual modalities for classification, routing and agent-action tasks based on predefined questions and answer choices. TechCrunch reported pricing of $0.10 per million input tokens, with no output-token charges. API product lead Nikunj Handa said the service reduced some response times from one to three seconds to 100, 120 or 180 milliseconds; these figures were not independently verified in the supplied reports. OpenAI reconfigured an existing Luna model for faster, lower-cost inference rather than training a separate decision model.
The release gives developers a fast, low-cost option for predefined classification and routing workflows and puts OpenAI in competition with TypeSafe’s Jev and other decision models.
OpenAI made its Decisions API model widely available to users for the first time, entering the market for fast, cheap intelligence that recently made TypeSafe, a startup founded by a former researcher at the frontier lab, a $7.6 billion company.
TypeSafe’s Jev model created a category for deep learning models that act as super classifiers: Given a set of options, they can quickly make choices and offer a confidence rating in their decision. Software developers quickly found that these decision or System One models could replace LLMs in a variety of workflows, operating more quickly and a at dramatically lower cost.
Nikunj Handa, the product lead for OpenAI’s APIs, told TechCrunch that his colleagues saw Jev and said “wow, we we have like a million use cases for this.” Handa’s team sprinted to offer their own version: Rather than building a specifically calibrated decision model, the team took ChatGPT Luna and reconfigured it to speed up the inference and lower the cost.
“The big unlock with Decisions API has been the speed, shrinking things from one to three seconds to 100, 120, 180 milliseconds,” Handa said. “In the future, we may sort of like have specific training to make our models like even better in these ways. But right now, it’s just plain old Luna that gets the job done.”