OpenAI Previews GPT‑5.6 Sol Ultrafast Mode

OpenAI has previewed its new mode called Ultrafast, which runs GPT-5.6 Sol at a much faster rate.

The Ultrafast mode makes GPT-5.6 Sol work at 14x the speed of Standard processing. It is powered by Cerebras for low-latency inference, generating up to 750 output tokens per second.

With GPT-5.6, OpenAI is expanding flagship model capabilities while improving efficiency across its entire infrastructure. These optimizations make high-tier AI capabilities significantly more cost-effective and accessible. While real-time speed previously required sacrificing model size or specialization, the new Ultrafast mode shifts focus toward maximizing useful work output per second.

The company is also testing the new mode in different scenarios:

• Incident Response & Reliability: Diagnoses system failures mid-outage by analyzing logs, recent code changes, and engineer reports to accelerate live fixes.
• Financial Research & Security: Detects fraud and tracks shifting market signals dynamically as conditions evolve.
• Customer Support & Voice: Resolves complex, multi-system queries in real time without pausing the conversation.
• Commerce: Answers inquiries, checks inventory, and clears checkout hurdles before potential buyers abandon their carts.
• Live Research & Experimentation: Converts overnight testing runs into continuous, interactive working sessions for real-time adjustments.

The ongoing preview relies on early customer feedback to pinpoint where high-speed performance adds the most value, helping shape and refine future product developments.

Additionally, internal teams at OpenAI are testing GPT-5.6 Sol on Ultrafast mode to accelerate real-time workflows while maintaining human oversight. The low-latency model speeds up IT incident response by instantly analyzing logs and trace data during active outages, while also compressing overnight research cycles into continuous, same-day testing iterations.

GPT‑5.6 Sol on Ultrafast mode is currently available in a limited preview to a select group of customers through the OpenAI API. OpenAI plans to expand access to more users soon.

ManilaShaker is a tech media producing insightful and helpful content for our local and growing international audience. Our goal is to create a premier Philippine digital consumer electronics resource that provides the most objective reviews and comparisons globally.