OpenAI has announced the global rollout of GPT-6 for ChatGPT, bringing its next generation of intelligence to over 1.2 billion weekly users. According to OpenAI, GPT-6 introduces Intelligent UI, a capability enabling the model to compose responses using text, graphics, tappable buttons, forms, charts, and interactive elements tailored to the user's query. A native component library and compiler allow interfaces to appear progressively as the model generates them. Additionally, GPT-6 can interleave thinking with answering, allowing it to begin responding to users while continuing to process information. In internal evaluations, GPT-6 Extra High begins answering in the same time as GPT-5.6 Medium while scoring higher, and GPT-6 Instant starts answering web search queries 44 percent sooner than its predecessor. Safety improvements include stronger resistance to multi-turn adversarial attacks, better handling of high-risk scenarios involving cyberattacks, biological threats, and violence, and clearer communication regarding model limitations. GPT-6 is powered by GPT-6 Sol for Plus, Pro, Business, and Enterprise tiers, and GPT-6 Luna for Free and Go tiers. The global rollout begins today for paid tiers and expands to Free and Go tiers tomorrow, though Enterprise availability depends on workplace admin settings. Models powering Work and Codex are not affected by this update.
GPT-6 and Intelligent UI for everyone
OpenAI is rolling out GPT-6 globally to more than 1.2 billion weekly ChatGPT users. The update introduces Intelligent UI, allowing the model to generate interactive interfaces alongside text, and enables the system to interleave thinking with answering for faster responses.
Why it matters
This release shifts ChatGPT from static text output to dynamic, context-aware interactive software interfaces generated on the fly. It also demonstrates progress in reducing latency through interleaved reasoning.
Topic dossiers OpenAI