GPT‑6.1 Sol Ultrafast mode: high-speed inference in the API and Codex
What is it?
Ultrafast mode, which runs GPT‑6.1 Sol up to 8x faster than standardVendor claim, is rolling out to the API, Codex and ChatGPT Work. It is called over WebSocket (an HTTP alternative exists). API pricing is $12 input and $60 output per million tokensVendor claim.
Why does it matter?
It is meant for work where waiting is the cost, like incident response, agents that operate apps directly, and real-time experiences. The per-token price is high, though. Routing only latency-sensitive steps to this mode makes more sense than sending every call through it.
Maturity
GAThe API is rolling out gradually across all supported regions, including US and EU data residency. In Codex and Work it is limited to Pro 500, usage-based Enterprise and credit-based Edu, and Enterprise admins have to turn it on.
How to track it
- Completion of the rollout across all regions and any wider plan access: check the OpenAI developer community thread.
- Changes to pricing or the call method (WebSocket, HTTP) and how the "up to 8x" speed holds up in practice: revisit the Ultrafast mode API docs.