What changed: OpenAI added Ultrafast mode for GPT-6.1 Sol in the Responses API. Using gpt-6.1-sol with service_tier: "ultrafast" reduces the time between generated output tokens. The feature is available to all API users subject to rate limits, with global processing and US and EU data residency options.
Who is affected
Businesses using the OpenAI Responses API, specifically the gpt-6.1-sol model.
What it means for your business
For businesses using gpt-6.1-sol through the Responses API, this means they can set service_tier to "ultrafast" to reduce the delay between output tokens, letting their applications respond faster.
What to do
If your Responses API integration uses gpt-6.1-sol, set service_tier to "ultrafast" and test the speed gain; review rate limits and the US/EU data residency options.
When it takes effect
Rolled out as of October 8, 2026.
Source: https://developers.openai.com/api/docs/changelog#2026-10-08
