The API Changelog Adds Ultrafast for GPT-6.1 Sol

The OpenAI API changelog entry for 8 October 2026 is labeled as a feature for the model gpt-6.1-sol on the Responses API, v1/responses. It adds Ultrafast mode. The instruction on the page is to use gpt-6.1-sol with service_tier set to "ultrafast" to reduce the time between generated output tokens. Availability is all API users, subject to rate limits, with global processing and US and EU data residency. The sentence ends by pointing to Ultrafast pricing. This changelog entry does not print a dollar amount, a token rate, or a comparison with the default tier.

The same October section has two nearby items that are not the Ultrafast switch. On 7 October the chat-latest snapshot was updated. The note says it points to the latest model available in ChatGPT for Plus, Pro, Business, and Enterprise users, and it recommends the GPT-6 model family for production API usage. On 6 October the Decisions API is listed in beta with gpt-6-luna, described as turning text and images into typed answers 10x faster than the Responses API. A second 6 October line says API usage tiers were simplified from five to three: Build, Launch, and Grow, with organizations upgrading as total credit purchases reach tier minimums.

What the Ultrafast line does not say

The Ultrafast paragraph does not name a context window, a maximum output length, or a region where the tier is refused beyond the rate-limit clause. Data residency is stated as global processing plus US and EU. Anyone pricing a call still has to open the pricing page the changelog cites. The parameter the note does specify is the pair of model name and service_tier value.

Source: OpenAI API changelog.