Signal

OpenAI previews 'Ultrafast' GPT-5.6 — 14x the speed, powered by Cerebras chips instead of GPUs

OpenAI is previewing an Ultrafast mode for GPT-5.6 Sol that runs up to 14 times faster than standard processing — around 750 output tokens a second — by routing inference through Cerebras hardware instead of GPUs, with no price or general release date confirmed yet.