OpenAI launched its Decisions API to all customers on 22 October. The company first showcased this technology during a developer event held on 29 September.
Response times for the Decisions API range between 100 and 180 milliseconds. By comparison, traditional AI response times typically last between 1 and 3 seconds.
OpenAI claims the new API can make decisions up to 10 times faster than the traditional Responses API. Input pricing for the service is set at $0.10 per million tokens, while cached input costs $0.01 per million tokens.
The API supports structured data, images, and text as inputs. Early adopters of the technology include Speak, Ramp, Datadog, and Box.
Andrew Hsu stated that a language-learning platform used the tool to cut the time between AI actions from 1.4 seconds to 0.13 seconds, an improvement of more than 90%. OpenAI believes that future updates could potentially lower response times to roughly 60 milliseconds.
Nikunj Handa suggested that these speeds might eventually enable developers to make AI decisions for individual frames in videos.
