Image: OpenAI
OpenAI is bringing GPT-6 to the main chat in ChatGPT, and this time free users are included. The update also adds a feature called Intelligent UI, which lets ChatGPT answer with charts, buttons, forms, maps and small working tools instead of only blocks of text.
ChatGPT just got a lot more visual—introducing Intelligent UI.
— ChatGPT (@ChatGPT) October 7, 2026
Intelligent UI allows ChatGPT to answer quickly with fully interactive user interfaces that make everyday answers more visual.
It makes learning complex topics easier, and it quickly creates tools to solve a task… pic.twitter.com/Fep4gvgQLM
Until now, GPT-6 sat off to the side. When OpenAI launched GPT-6 Sol and GPT-6 Luna last month, it said the models were "not yet available in Chat," only in ChatGPT Work and Codex. This release brings it to the regular Chat tab. OpenAI says more than 1.2 billion people use ChatGPT each week.
The rollout starts globally with Plus, Pro, Business and Enterprise, per OpenAI's announcement, and expands to Free and Go a day later. Enterprise access depends on workspace admin settings.
"For everyone" doesn't mean everyone gets the same model, though. Paid plans run on GPT-6 Sol, while Free and Go get GPT-6 Luna, according to OpenAI's release notes. OpenAI says both are tuned for everyday conversation, but they sit at very different price points. In the API, Luna lists at $0.10 per million input tokens and $0.50 per million output tokens, against $2 and $10 for Sol. That's a 20x gap. OpenAI's announcement also doesn't say how much GPT-6 free users get before hitting usage limits.
There's a catch at the top end too. The release notes say the Pro reasoning option keeps using GPT-6 Astra, OpenAI's most capable model, and doesn't support Intelligent UI. Intelligent UI works from Instant through Extra High reasoning. The models behind Work and Codex aren't changing either, so GPT-6.1 Sol, which OpenAI launched at DevDay for those areas, isn't part of this update.
So what does Intelligent UI actually do? OpenAI says it trained GPT-6 to decide how to mix text, visuals and interactive pieces based on the question. Its examples include cooking timings shown alongside a Sunday roast recipe, road-trip stops on a map, a clickable diagram of a bike's parts, and a bill splitter or savings calculator built on request. When plain text is the better answer, it's supposed to stick with text.
Behind the scenes, OpenAI says it built a library of ready-made interface parts plus a compiler that lets the layout appear piece by piece as the model writes, rather than after the whole reply is done. The company admits there's "still work ahead to improve the model's design judgment."
GPT-6 can also start answering while it's still thinking, then keep adding to the reply. OpenAI says GPT-6 Instant starts answering web-search questions 44% sooner on average than GPT-5.6 Instant. It also says GPT-6 Extra High begins answering as quickly as GPT-5.6 Medium while scoring better than GPT-5.6 Extra High on everyday agent-style tasks. Both claims come from OpenAI's internal evaluations and haven't been independently tested.
On safety, OpenAI says GPT-6 is better at resisting attempts to get around its safety training, especially ones spread across several messages, and better at saying when it can't answer something. The details are in its system card.
The demos look polished, but they're OpenAI's own. The real test is whether interactive answers hold up on ordinary questions once Free and Go users are on Luna.
Comments