Intel / article

Previewing Ultrafast mode: GPT-5.6 Sol at up to 14X the speed

Source: openai.com
Date: Published Aug 13, 2026

Context

Ultrafast is a service tier for running GPT-5.6 Sol more quickly. In its preview announcement, OpenAI reports up to 750 output per second and up to fourteen times the speed of Standard processing. The change concerns serving the model, so it addresses how quickly generated text becomes available.

The preview initially involved a select group of customers. Its headline figures are maximum reported speeds for output generation, not a promise that every complete workflow finishes fourteen times sooner.

Terms in this piece · Glossary

token — The chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.

Read the source openai.com