OpenAI Launches Ultra-Fast Mode for GPT-5.6 Sol at 750 Tokens per Second
On August 13, 2026, OpenAI introduced an ultra-fast mode for GPT-5.6 Sol, delivering output speeds up to 750 tokens per second—14 times faster than the standard mode. Powered by Cerebras' wafer-scale engine with 44GB of on-chip SRAM, the mode outperforms Fable 5 by 11 times and Opus 4.8 by 5 times. In the Human Ultimate Exam, it processed 2,500 questions in 11 hours 11 minutes versus Fable 5's 78 hours 27 minutes, with comparable accuracy. Access is initially limited to select customers, evaluated by workload fit and compute availability.
On August 13, 2026, OpenAI officially launched the "ultra-fast" mode for GPT-5.6 Sol, which delivers output speeds of up to 750 tokens per second—14 times faster than the standard mode. OpenAI first mentioned in June 2026, when it unveiled the GPT-5.6 series, that the Sol version would arrive on the Cerebras platform in July, but no specific details were disclosed throughout that month. Today's release marks the mode's formal availability to a select group of customers.
The mode is powered by Cerebras' wafer-scale engine architecture, which uses 44GB of on-chip static random-access memory to alleviate data transfer bottlenecks. Performance data shows it is 11 times faster than Fable 5 and 5 times faster than Opus 4.8. In the Human Ultimate Exam, the Sol ultra-fast mode processed 2,500 questions in 11 hours and 11 minutes, while Fable 5 took 78 hours and 27 minutes, with comparable accuracy.
The mode is applicable to event response, financial research, real-time customer support, business, programming, and research workflows. Due to limited Cerebras compute capacity, OpenAI will evaluate access based on customer workload fit and compute availability. Customers can submit a request form to receive notifications.
Why this event matters
The event has a measured impact on 1 industry. The strongest current signal is positive for Artificial Intelligence, with intensity 55/100 and 80% confidence over a short term horizon.
Impact figures are analytical estimates that combine direction, intensity, confidence and event importance. They are not investment advice.