OpenAI's new 'Ultrafast' mode runs GPT-5.6 Sol at 14x speed

OpenAI has rolled out a new mode called Ultrafast, designed to significantly accelerate how fast its latest and most powerful model, GPT-5.6 Sol, completes its work. The company says Ultrafast can operate at 14 times the speed of standard processing, delivering up to 750 output tokens — the distinct pieces of text an LLM generates while interacting with a user — per second.

“Until now, getting real-time speed typically meant choosing a smaller or more specialized model,” OpenAI said in a blog post Thursday. “Ultrafast points to progress in a new direction: more useful work per second.” OpenAI’s competitors, including Anthropic, have similarly launched accelerated versions of their models; Claude has a fast mode, though OpenAI says it doesn’t deliver the kind of speed Ultrafast offers.

OpenAI suggests this high-octane version of GPT-5.6 Sol can be deployed across a range of corporate workflows, most notably incident response, customer service and support, financial market analysis, and e-commerce, among other areas. Ultrafast, currently released in preview, is powered by OpenAI’s partnership with chipmaker Cerebras. That preview is only available to a small group of customers for now, though OpenAI says it will expand access as “capacity grows.”

Why it matters

Pairing a frontier model with specialized inference hardware to hit real-time speeds — instead of shrinking the model — suggests OpenAI sees latency, not just capability, as the next competitive front in enterprise AI.