Noesion Radarnoesion.ai
Release2026-08-10

OpenAI previewed Ultrafast, a new speed setting in its developer service that runs the GPT-5.6 Sol model up to 14 times faster than normal, using hardware from Cerebras. It is aimed at work where a reply has to arrive during a live conversation or an unfolding incident. Access is limited to a selected group of customers for now, with a signup list for later.

What changed

Real-time response speeds generally required choosing a smaller or more specialized model.

What it unlocks

Running OpenAI's most capable model inside live interactions such as voice calls, checkout help and unfolding incidents.

  • up to 14× faster than standard
  • up to 750 output tokens per second

What you need to act on it

  • OpenAI API access
  • selection into the limited preview
  • waitlist signup

Sources

Send this to someone who needs it

Shares the story and its sources. Nothing about you.

What does this mean for your job?

This is the story as everyone gets it. Once a week we send you the version written for your role — what changed, why it matters for the work you actually do, and one thing to try. Free while we tune it.