OpenAI Launches GPT-6.1 Ultrafast Mode, Generation Speed Up to 8x Faster

en.Wedoany.com Reported - On October 8, U.S. artificial intelligence company OpenAI officially launched the Ultrafast mode for GPT-6.1 Sol, covering OpenAI API, Codex, and ChatGPT Work. This mode delivers token generation speeds up to 8 times that of the standard mode, with API pricing at $12 per million input tokens and $60 per million output tokens, both 6 times the standard mode price. The new mode is primarily aimed at tasks such as software debugging, agent-operated applications, and low-latency interactions.

On the API side, developers can call the model gpt-6.1-sol via the Responses API and set service_tier to ultrafast to enable Ultrafast mode. This mode supports WebSocket connections and also offers HTTP invocation. OpenAI's developer documentation shows that Ultrafast mode is available to all API users, but actual calls remain subject to rate limits.

In Codex and ChatGPT Work, Ultrafast mode is currently available to Pro 500 subscribers, eligible usage-based Enterprise users, and Education users on credit-based billing. Enterprise administrators need to enable the corresponding access permissions for their organization. This rollout covers all supported regions, with U.S. and EU data residency support. OpenAI has also simultaneously added EU data residency support for GPT-6.1 Sol Fast and GPT-6 Luna Fast.

GPT-6.1 Sol was released on September 29 as an upgraded version of GPT-6 Sol, with standard API pricing at $2 per million input tokens, $10 per million output tokens, and $0.10 per million cached input tokens. OpenAI positions it for complex programming, computer operation, and professional work tasks, supporting a context window of up to 1.05 million tokens and a maximum output of 128,000 tokens per request. This Ultrafast mode uses the same model identifier, differentiating processing speed and billing standards through service tiers.