OpenAI Previews 750-Tokens-per-Second GPT-5.6 Sol Tier on Cerebras Hardware

OpenAI previewed Ultrafast, a new API tier for GPT-5.6 Sol delivering up to 750 output tokens per second — roughly 14 times standard speed. The tier runs on Cerebras wafer-scale inference chips rather than OpenAI's own infrastructure. Access is limited to a small customer group, with expansion planned as capacity grows. Pricing and general availability remain undisclosed.