News in Short
- OpenAI has introduced an early preview of Ultrafast mode for GPT-5.6 Sol.
- The new mode is powered by Cerebras and can generate up to 750 output tokens per second.
- OpenAI claims GPT-5.6 Sol can run up to 14 times faster than standard processing.
- The company is testing Ultrafast with businesses across coding, commerce, financial research and customer support.
OpenAI has introduced an early preview of Ultrafast, a new processing tier designed to make GPT-5.6 Sol significantly faster. The company says the mode can deliver up to 750 output tokens per second.
GPT-5.6 Sol Gets a Speed Boost
OpenAI announced Ultrafast through a blog post as a limited preview for select customers. The new API service tier focuses on reducing the time GPT-5.6 Sol takes to complete tasks. The company says the higher processing speed could help businesses build more responsive AI products. It could also help teams make faster decisions and use AI in time-sensitive workflows.
OpenAI has partnered with Cerebras to power GPT-5.6 Sol in Ultrafast mode. The company claims the model can generate up to 750 output tokens per second.
Built for Real-Time AI Workflows
OpenAI is testing Ultrafast across several business applications. These include coding, commerce, financial research, customer support and other interactive workloads. The company has highlighted use cases such as real-time incident response, market-signal analysis, transaction assessment, customer support and live research.
OpenAI is also testing Ultrafast internally. Its teams are using the mode for incident response and research workflows involving search, connected tools and information gathering. For now, access remains limited to a select group of customers. OpenAI says it plans to expand access as more capacity becomes available.
