UpdateOpenAI · Announced 2026-10-08 · Found by us 10/13/2026

GPT-6.1 Sol adds Ultrafast mode in the Responses API

GPT-6.1 Sol has a new Ultrafast mode that reduces the time between generated output tokens. It is available to all API users, with US and EU data residency options.

What changed

OpenAI added Ultrafast mode for GPT-6.1 Sol in the Responses API. Developers enable it by using gpt-6.1-sol with service_tier set to "ultrafast". The mode supports global processing and US and EU data residency.

Why it may matter for your work

Apps using Sol could display generated answers more quickly once output begins. Businesses needing US or EU data residency may also be able to use this option.

What it can actually do

Reduce the time between generated output tokens in GPT-6.1 Sol API responses.

Limits to check

Subject to rate limits. The page links to separate Ultrafast pricing but does not state the rates.

“It is available to all API users, subject to rate limits, with global processing and US and EU data residency.”From the linked source page