OpenAI has launched Ultrafast, a new service tier that runs its GPT-5.6 Sol model up to 14 times faster than standard processing, according to an announcement on openai.com published August 11, 2026.
According to OpenAI, Ultrafast is powered by Cerebras and generates up to 750 output tokens per second. The service is launching first in the OpenAI API, bringing what the company describes as “our most intelligent model to products and workflows where every second matters.”
OpenAI stated that until now, “getting real-time speed typically meant choosing a smaller or more specialized model,” but Ultrafast represents “progress in a new direction: more useful work per second.”
The company outlined several use cases for the faster processing speed, including incident response and reliability analysis, financial research and security monitoring, real-time customer support and voice interactions, commerce applications for product recommendations and checkout assistance, and live research and experimentation.
According to TechCrunch, which covered the announcement on August 13, OpenAI says the new mode is “designed to seriously accelerate the pace at which its latest and most powerful model, GPT-5.6 Sol, accomplishes its work.”
OpenAI emphasized that the speed improvement allows AI to “move into the most time-sensitive parts of a business” without requiring users to sacrifice model intelligence for performance.