POPA ReconHow it investigates

OpenAI Ultrafast mode for GPT-6.1 Sol (announced 2026-10-08): what it is, where it runs, what it costs, who can use it.

OpenAI

OpenAI Ultrafast mode for GPT-6.1 Sol (announced 2026-10-08): what it is, where it runs, what it costs, who can use it

Ultrafast is rolling out for GPT-6.1 Sol in the API, Codex and ChatGPT Work, with "near-Astra intelligence at up to 8x faster speeds than Sol Standard", per the OpenAIDevs announcement. API pricing is $12 per million input tokens and $60 per million output tokens. In Codex and Work, access needs a Pro 500 plan or an eligible Enterprise or Edu plan.

Speed

OpenAI states "up to 8x" faster than Sol Standard. The figure is a ceiling, not a guaranteed rate. The docs page calls Ultrafast the fastest service tier in the OpenAI API. It recommends WebSockets, especially for agentic apps that make many tool calls in quick succession, because without a persistent connection network overhead can reduce the latency gains. HTTP also works: set service_tier: "ultrafast" in the Responses API.

Pricing

OpenAI's pricing post gives the $12 and $60 per million token rates. It positions the mode for debugging an outage, agents navigating apps, and live experiences. A commenter, @TokenGremlin, put GPT-6 Astra Ultrafast at $60 input and $300 output per million tokens. @santtiagom_ calls the Sol rate 6x more expensive in the API; this is that user's claim, not OpenAI's. The docs point to an Ultrafast pricing table for cached input and cache write prices, which the text read here does not show.

Availability and access limits

  • API: open to API users, with these default Ultrafast token limits for GPT-6.1 Sol per the docs:

API usage tierTokens per minute
Build1,000,000
Launch4,000,000
Grow40,000,000

  • Codex and ChatGPT Work: per OpenAIDevs, available on Pro 500, eligible usage-based Enterprise, and credit-based Edu plans. Enterprise admins must enable access.

  • Regions: all supported regions. GPT-6.1 Sol Ultrafast supports US and EU data residency and global processing. OpenAI also added EU data residency for GPT-6.1 Sol Fast and GPT-6 Luna Fast. GPT-6 Astra Ultrafast supports US data residency and global processing only.

  • Other models: the docs say Ultrafast is broadly available for GPT-6 Astra and GPT-6.1 Sol, with preview access for GPT-5.6 Sol.

  • Rollout: the posts say "rolling out", so it may not yet appear for every account.

Pages: Announcement · Pricing post · Access post · Ultrafast docs

Ask your own