14 times faster

14 times faster! OpenAI has launched the GPT-5.6 Sol UltraFast ultra-fast mode, which can reach up to 750 tokens per second. OpenAI has officially launched a preview of the new GPT-5.6Sol UltraFast ultra-fast mode for enterprise users. With this mode, the running speed of the model has been significantly improved, up to 14 times that of the standard mode, and up to 750 tokens can be generated per second. It is understood that the core technology of this ultra-fast mode is strongly supported by Cerebras. However, in order to ensure the rational use of resources, this model currently adopts an application review system. Enterprise users in need need to submit specific usage scenario applications to OpenAI. Whether they are approved will be comprehensively evaluated by the official based on the actual situation. In terms of application scenarios, OpenAI said that the ultra-fast mode is not suitable for all conventional workloads, but is specially designed for business scenarios that have extremely high real-time requirements. For example: real-time voice interaction, intelligent customer support, business applications, development of agents (Agents), in-depth financial research, and cutting-edge security research, etc. As the preview phase progresses, the computing resources provided by Cerebras will continue to be optimized. In the future, OpenAI plans to gradually expand the coverage based on feedback from enterprise workloads, allowing more enterprises to smoothly call this extremely fast AI capability through API interfaces. via AI News (author: AI Base)