Faster Inference, More Interaction
NVIDIA reported that GPT-6 Astra Ultrafast, running on NVIDIA Blackwell GPUs, is available in the OpenAI API and to eligible ChatGPT Work and Codex users. The report describes the model as using inference optimizations from OpenAI models.
NVIDIA reported that GPT-6 Astra Ultrafast, running on NVIDIA Blackwell GPUs, is available in the OpenAI API and to eligible ChatGPT Work and Codex users. The report describes the model as using inference optimizations from OpenAI models.
The primary source was published October 1, 2026, and states that availability is current for the listed OpenAI API and eligible ChatGPT Work and Codex users.
NVIDIA reports that OpenAI’s GPT-6 Astra Ultrafast is now available in the OpenAI API and to eligible ChatGPT Work and Codex users, running on NVIDIA Blackwell GPUs. How could faster inference change the design of digital workers, game systems, or interactive 3D experiences? martinezaistudios.com