If you’ve ever paused mid-chat because ChatGPT seemed slow, OpenAI has your back with its latest innovation: Ultrafast mode for GPT-5.6 Sol.

This turbocharged version of the model can process text at an impressive 14 times the speed of standard operations—cranking out up to 750 output tokens per second. For context, that’s akin to generating about three full paragraphs in a fraction of a second.

OpenAI suggests this ultra-speedy mode could transform corporate workflows ranging from customer service to financial analysis, but it's currently only available to a select few users as the company scales up capacity.

The move mirrors similar efforts by rivals like Anthropic, which now offers its Claude model in a “fast” version. However, OpenAI’s offering is significantly faster and more versatile, potentially setting new benchmarks for speed in AI.