344
Productivity & Workflow355
Automation & Workflow224
Software Development251
Marketing & Growth192
AI Infrastructure & MLOps174
Writing & Content Creation203
Data & Analytics142
Photography & Imaging156
Design & Creative170
Customer Support132
Sales & Outreach125
Voice & Speech135
Education & Learning131
Operations & Admin87
OpenAI is testing Ultrafast, a new mode that it says makes GPT 5.6 Sol respond much faster, aimed at business uses like support and incident response.
In short: OpenAI has launched a preview of a new “Ultrafast” mode that it says makes its GPT 5.6 Sol model run up to 14 times faster.
OpenAI announced a new mode called Ultrafast for GPT 5.6 Sol, which the company describes as its latest and most powerful model. A “mode” here is like switching your car from normal driving to a sport setting, with the goal of getting answers sooner.
OpenAI says Ultrafast can generate up to 750 “tokens” per second. Tokens are small chunks of text that an AI uses to produce an answer, similar to how a sentence is built from words and pieces of words.
The company says that, in the past, getting real time speed often meant using a smaller model. With Ultrafast, OpenAI says it can do more work per second without switching to a simpler system.
OpenAI also said Ultrafast is powered by its partnership with Cerebras, a company that makes AI computer chips. The preview is only available to a small group of customers right now, and OpenAI says access will expand as it has more capacity.
Competitors are also working on faster options. For example, Anthropic offers a fast mode for its Claude model, according to the report.
Faster responses can matter most in business settings where time is tight, like customer support chats, monitoring problems in online systems, or tracking fast moving financial markets. For regular people, these speedups could show up as quicker support experiences and faster tools at work, but availability is limited for now.
Source: TechCrunch AI