BusinessFriday, August 14, 2026· 2 min read

OpenAI Previews Ultrafast GPT-5.6 Sol at Up to 14× Speed

Source: OpenAI Blog

TL;DR

OpenAI is previewing a new Ultrafast API service tier that can run GPT-5.6 Sol up to 14 times faster. Powered by Cerebras, the tier reaches up to 750 output tokens per second, opening the door to more responsive AI apps and real-time workflows.

Key Takeaways

  • 1OpenAI’s new Ultrafast tier is designed to dramatically reduce latency for API users.
  • 2GPT-5.6 Sol can run up to 14× faster through the new service tier.
  • 3The system is powered by Cerebras and can deliver up to 750 output tokens per second.
  • 4Faster generation could improve real-time assistants, coding tools, customer support, and interactive AI products.

OpenAI has previewed Ultrafast, a new API service tier built to make GPT-5.6 Sol significantly faster for developers and businesses. The company says the tier can run the model at up to 14× the speed, marking a major step toward more responsive AI experiences.

Powered by Cerebras, Ultrafast can deliver up to 750 output tokens per second. That kind of throughput could make AI systems feel far more immediate in practical use cases, from live coding copilots to real-time customer support and interactive creative tools.

Why it matters

Speed is one of the biggest factors shaping how useful AI feels in everyday products. By reducing wait times, faster inference can help developers build AI applications that are smoother, more conversational, and better suited to time-sensitive workflows.

  • For developers: faster responses can improve product design and user engagement.
  • For businesses: lower latency can support higher-volume, real-time AI services.
  • For users: AI tools can feel more natural, responsive, and productive.
Read the original article

More in Business

Business

Databricks Raises $5B as Investor Demand Surges for AI Data Infrastructure

Databricks has secured a massive $5 billion funding round at a reported $190 billion valuation, reflecting strong investor confidence in the company’s role powering AI development. The raise gives Databricks more resources to scale the data and AI infrastructure businesses rely on to build, deploy, and manage intelligent systems.

Business

Writer Debuts Lower-Cost Enterprise AI Model to Make Deployment More Affordable

Writer has introduced a new AI model, built as a post-training variation of Z.ai’s open source GLM-5.2, aimed at delivering deployment-ready enterprise capabilities at a lower price. The company also upgraded its harness to help customers better contain token costs, making large-scale AI adoption more practical for businesses.

Business

OpenAI’s GPT-5.6 Guide Helps Startups Build Faster, Smarter AI Agents

OpenAI’s new builder guide shows startups how to use GPT-5.6 to create AI agents more quickly and cost-effectively. With smarter model selection and expanded Responses API capabilities, teams can ship more capable products while optimizing performance and spend.

Get AI Wins in Your Inbox

The best positive AI stories delivered to your inbox. No spam, unsubscribe anytime.