> ## Content Index
> Fetch the complete content index at: https://www.testingcatalog.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# OpenAI prepares to expand Ultrafast API to more users
- URL: https://www.testingcatalog.com/openai-prepares-to-expand-ultrafast-api-to-more-users/
- Published: 2026-09-26T12:24:22.000Z
- Updated: 2026-09-26T12:24:22.000Z
- Description: What do we know so far? Ultrafast mode is coming to OpenAI Playground and will likely become broadly accessible after the OpenAI DevDay.
- Author: Alexey Shabanov
- Tags: OpenAI, Latest AI News, AI Rumours

OpenAI appears to be preparing a wider rollout of its Ultrafast API mode around DevDay on September 29, with new references showing up across the OpenAI Platform and API documentation.

TestingCatalog spotted a dedicated speed selector being prepared for the Responses API Playground (currently hidden), where developers could choose between Standard, Fast, and Ultrafast processing. OpenAI has already officially previewed Ultrafast with GPT-5.6 Sol, describing speeds of up to 750 output tokens per second and up to 14× faster inference than Standard. Access remains limited to selected customers, and OpenAI confirmed that the mode is powered by Cerebras.

> OpenAI just added “Ultrafast” to its cloud-agent API docs 👀[https://t.co/s6u44II4D6](https://t.co/s6u44II4D6?ref=testingcatalog.com) [pic.twitter.com/6LX0IkwQeS](https://t.co/6LX0IkwQeS?ref=testingcatalog.com)
> 
> — imjustnewatai (@imjustnewatai) [September 26, 2026](https://x.com/imjustnewatai/status/2103752858734858568?ref%5Fsrc=twsrc%5Etfw&ref=testingcatalog.com)

The timing makes broader availability during DevDay plausible. OpenAI has since released the GPT-6 family, including [GPT-6 Sol](https://www.testingcatalog.com/openai-launches-faster-cheaper-gpt-6-sol-and-luna/) and [GPT-6 Astra](https://www.testingcatalog.com/openai-launches-gpt-6-astra-across-chatgpt-and-api/), making support for these newer models a key thing to watch. No one has confirmed that every GPT-6 model will support Ultrafast at launch.

For developers, the main trade-off will likely be economics. Standard, Fast, and Ultrafast could let organizations choose latency based on each workload's value, reserving higher-cost inference for applications where response time directly affects revenue or productivity.

The feature also fits [OpenAI’s](https://www.testingcatalog.com/tag/openai/) broader infrastructure strategy. The company announced a 750 MW Cerebras partnership earlier this year, with capacity being deployed in stages through 2028\. TestingCatalog has separately spotted new [OpenAI Platform onboarding tiers](https://www.testingcatalog.com/devday-new-plans-and-new-platform-for-building-ai-apps/), including an Accelerate option aimed at production workloads. Together, these changes suggest DevDay could focus heavily on API infrastructure, compute tiers, and tools for developers building production AI systems.