GPT-6 Brings Intelligent UI to ChatGPT
OpenAI began rolling out GPT-6 with Intelligent UI in ChatGPT, generating dynamic charts, calculators, and interactive forms directly in chat.
The update replaces plain text responses with on-demand interactive charts, calculators, and input forms.
ChatGPT is moving away from conversations made strictly of plain text. OpenAI has started rolling out GPT-6 alongside Intelligent UI, an interface system that renders visual widgets and interactive micro-tools directly in the conversation window for more than 1.2 billion weekly users. The company announced the update on its official blog and outlined rollout details in support documentation.
Instead of producing static Markdown tables or raw blocks of code, the system generates functional software components inside the stream. When users request a budget breakdown, an expense split, or spreadsheet analysis, the model responds with interactive charts, clickable buttons, and operational calculators. In an interview with Wired, product manager Aarush Selvan noted that standard text outputs often created friction for practical, daily tasks.
Real-time Intelligent UI compilation
Under the hood, the system couples a library of native components with a real-time compiler. The model does not need to complete its entire reasoning chain before drawing components on screen. Interface code streams and renders progressively as explanatory text appears.
This continuous rendering runs alongside interleaved reasoning. For queries requiring web searches, OpenAI stated that the GPT-6 Instant variant displays initial results 44% faster than GPT-5.6 Instant, cutting down the pause before the first usable interaction.
Tiered release schedule and missing API
The rollout follows a split schedule. On October 7, 2026, paid tiers including Plus, Pro, Business, and Enterprise gained access to the mid-tier GPT-6 Sol model. On October 8, Free and Go tier accounts received GPT-6 Luna, a lighter version built for lower latency, as confirmed via OpenAI’s official account on X.
For now, the dynamic interface layer remains exclusive to ChatGPT’s first-party web and mobile apps. OpenAI did not release an SDK, documentation, or public API endpoints for third-party developers looking to implement similar streaming components in external software.
Technical limits and turning it off
Not every mode in ChatGPT supports the interactive components. Intelligent UI operates across Instant through Extra High reasoning settings, but stays disabled in Pro reasoning mode powered by GPT-6 Astra. Voice conversations and dedicated tabs like Work and Codex also remain purely text or voice. Component state is preserved only within an active chat; closing the thread or opening another one wipes the temporary inputs.
Users who prefer standard text responses can disable visual components entirely. The setting is located under Personalization → Layout and Visuals → Simple, and can also be enforced through account Custom Instructions.