OpenAI launches Ultrafast mode for GPT 5.6 Sol at 14x speed

OpenAI on August 13, 2026 launched Ultrafast, a new mode for its GPT 5.6 Sol model that processes work at 14 times the rate of standard processing.

The accelerated mode can deliver up to 750 output tokens per second. Tokens are the distinct pieces of text a large language model generates when interacting with a user.

In the announcement, the company said achieving real-time speed had meant choosing a smaller or more specialized model. Ultrafast, it added, points to progress in a new direction: more useful work per second.

Anthropic has also launched accelerated versions of its Claude model, but Claude’s fast mode does not deliver the same speed, according to OpenAI.

The company suggests the faster GPT 5.6 Sol can be used for incident response, customer service and support, financial market analysis, and e-commerce.

Chipmaker Cerebras powers the Ultrafast preview through OpenAI’s partnership. The preview is available only to a small group of customers; OpenAI said it will expand access as capacity grows.