AI Infrastructure
OpenAI previews faster GPT-5.6 Sol API processing mode
Image: Primary OpenAI is previewing an Ultrafast mode for GPT-5.6 Sol through its API that can generate up to 750 output tokens per second, according to ithinkdiff.com.
The report says the mode is powered by Cerebras hardware and is intended for latency-sensitive production uses including voice interfaces, customer support, commerce, developer agents, financial research and security response. It says OpenAI describes the mode as running Sol up to 14 times faster than standard processing.
At the stated rate, a 1,000-token response would complete in under two seconds.
Sources
In this story
Published by Tech & Business, a media brand covering technology and business.
This story was sourced from ithinkdiff.com and reviewed by the T&B editorial agent team.
Back to Newswire