0:00 / 0:00
ai
Previewing Ultrafast mode: GPT‑5.6 Sol at up to 14X the speed
OpenAI14 August 2026Watch on YouTube
Description
Today we’re previewing Ultrafast mode, a new service tier in the OpenAI API for GPT‑5.6 Sol. Powered by Cerebras, Ultrafast runs up to 14x faster than Standard processing and generates up to 750 output tokens per second. OpenAI technical staff are already seeing what that speed changes. One security investigation workflow that once took one to two hours now takes 10–15 minutes, sometimes approaching real time. In other workflows, staff use Ultrafast to investigate root causes, search systems in parallel, and stay in flow while coding. Ultrafast is available to a select group of API customers, with access expanding as capacity grows. Learn more: https://openai.com/index/previewing-ultrafast/
What you'll learn
- Ultrafast mode is a new API tier for GPT-5.6 Sol that runs up to 14x faster than standard processing
- The Ultrafast version generates up to 750 output tokens per second, powered by Cerebras hardware
- Security investigations that previously took 1-2 hours now complete in 10-15 minutes
- Ultrafast enables parallel searching and real-time workflows for root cause investigation and coding
Frequently asked questions
What is Ultrafast mode and how much faster is it than standard?
Ultrafast mode is a new API tier for GPT-5.6 Sol that runs up to 14x faster than standard processing. Powered by Cerebras hardware, it generates up to 750 output tokens per second.
What practical benefits does Ultrafast provide for security investigations?
Security investigations that previously took 1-2 hours can now be completed in 10-15 minutes. This enables teams to investigate root causes faster and sometimes achieve real-time results.
Who has access to Ultrafast mode?
Ultrafast is currently available to a select group of OpenAI API customers. Access expands as capacity grows.