ChatGPT GPT Astra: Nvidia CEO Jensen Huang Praises OpenAI Model and Declares AGI Has Arrived

Nvidia chief executive Jensen Huang has said artificial general intelligence has arrived, crediting OpenAI’s newly released GPT-6 Astra model. In a post on X he congratulated the OpenAI team and highlighted the scale of computing power used to train the system.

Huang wrote that GPT-6 Astra was trained on roughly 100,000 or more Nvidia Grace Blackwell NVLink72 systems. He framed the development as rapid progress “from ChatGPT to o1 to Astra in 4 years” and added that a further 400,000 GPUs are coming online next. OpenAI GPT-6 Astra Model Launched With Advanced Capabilities.

'We Are in AGI Era Now', Says OpenAI President Greg Brockman

GPT-6 Astra Capabilities and Rollout

OpenAI launched GPT-6 Astra in early September 2026, describing it as its most intelligent and aligned model to date. The company said the model is state-of-the-art in computer use, browsing, software engineering, cybersecurity, science and professional work.

It has posted high scores on key benchmarks, including 98 per cent on FrontierMath Tier 4, 99.9 per cent on ARC-AGI-3 and 100 per cent on ExploitBench. Astra is designed to handle longer and more complex tasks with less human guidance, including operating computers on a user’s behalf.

The model is becoming available to ChatGPT Plus, Pro, Business and Enterprise users, as well as through the OpenAI API and cloud partners. Rollout began with limited organisations and is expanding to wider paid tiers.

OpenAI Response on the AGI Era

OpenAI president and co-founder Greg Brockman replied to Huang’s comments. He said the company is now moving into the AGI era, whether that is viewed as applying to the current model, the previous one or the next, and that the progress would not have been possible without close partners.

OpenAI has positioned Astra as a generational step that could later be seen as marking the start of the AGI period. The firm has also stressed additional work on safety and alignment before the wider release.

Nvidia Computing Scale Behind Astra

Huang’s remarks underline the heavy reliance on advanced Nvidia hardware for frontier model training. The reference to more than 100,000 Grace Blackwell systems and the planned addition of hundreds of thousands more GPUs points to the continuing expansion of AI infrastructure needed to support such models. SpaceX Trying to Resolve OpenAI’s Cutoff of Cursor Platform for Users.

The exchange between the two executives highlights both the technical milestone claimed for GPT-6 Astra and the close partnership between OpenAI and Nvidia in the current phase of AI development.