Back to rankings

GPT-6.1 Sol Ultrafast Mode Speeds Up Codex Generation

CodexView original changelog

OpenAI introduced Ultrafast mode for GPT-6.1 Sol in Codex and ChatGPT Work, speeding up token generation for the model. Ultrafast is available on the Pro $500 plan and eligible Enterprise and Edu plans. Enterprise access is off by default and must be enabled by workspace owners. The mode supports inference residency in the United States and Europe.

Key Takeaways

  • Ultrafast mode speeds up token generation for GPT-6.1 Sol in both Codex and ChatGPT Work.
  • It is gated to the Pro $500 plan and eligible Enterprise and Edu plans, so it is not available to every Codex user.
  • Enterprise access is off by default, so workspace owners have to enable it explicitly.
  • The tier supports inference residency in the United States and Europe (EEA plus Switzerland), which matters for regulated teams.
  • It builds on GPT-6.1 Sol, which became the default Codex coding model on September 29.
  • Usage is billed at separate credit rates, so heavy users should check the pricing page first.

Faster generation for GPT-6.1 Sol

OpenAI added Ultrafast mode to GPT-6.1 Sol in Codex and ChatGPT Work. The mode speeds up token generation for the model, which shortens the wait during long agentic coding runs, large refactors and multi-step tasks where output volume is high.

Ultrafast arrives after GPT-6.1 Sol became the default coding model in Codex on September 29. It is a premium speed tier on top of that model, not a different model.

Who can use it

Ultrafast is available on the Pro $500 plan and on eligible Enterprise and Edu plans. For Enterprise workspaces, access is off by default. Workspace owners must turn it on before members can select it, so teams on managed plans should check with their admin if the option does not appear.

Data residency

GPT-6.1 Sol Ultrafast supports inference residency in the United States and in Europe (the EEA plus Switzerland). Organizations with regional processing requirements can therefore use the faster tier without giving up residency commitments.

Where to find details

OpenAI's entry points to separate documentation covering the plan and workspace requirements for Ultrafast mode, model guidance for GPT-6.1 Sol, and the pricing page that lists usage and credit rates. Because Ultrafast is a paid speed tier, the credit rates are worth reviewing before switching it on for heavy workloads.