ChatGPT answers can now be a game or a budget calculator
A ChatGPT answer can now be a chart, a form or a working calculator, not only a block of text. OpenAI announced the feature, Intelligent UI, in an October 7 post on X. It comes with GPT-6 in the Chat tab: Plus, Pro, Business and Enterprise users get it globally that day, and Free and Go users start getting it the next day.
OpenAI says GPT-6 builds each response out of text, visuals and interactive elements, “choosing how they fit together based on your question.” Those elements include graphics and charts, tappable buttons, forms and small tools that run inside the conversation. The company’s examples are a game and an interactive budget planning calculator.
This is the second step of the GPT-6 rollout. GPT-6 replaced GPT-5.6 as the model behind ChatGPT for paid users in September, Mashable reported, but that rollout did not include Intelligent UI. According to the outlet, OpenAI built a library of components the model can draw from when it creates interactive visual content, then trained the model to use them for different prompts. OpenAI’s own description of the result: “GPT‑6 learned to use the component library and make good design decisions, including how to organize information clearly, when to use interactivity, and when a simple text response is enough.”
So plain text is still an option. It is the answer ChatGPT gives when it suits the question best. The OpenAI examples Mashable lists show the range: an interactive diagram of a bicycle, a graph explaining GDP, a bill-splitting calculator and a retirement-savings projection tool.
Something similar already exists in a smaller form. OpenAI’s documentation describes a Visualizations preview that turns questions into charts, maps, diagrams, calculators, simulations and interactive explanations. ChatGPT picks a visual format there when it materially improves the answer, and users can tag @Visualize to ask for one. Whether you get it depends on your plan, platform, account and workspace settings. The three examples in those docs, a spirograph, a wave interference lab and a GPT tokenizer explainer, come from the GPT-5.6 launch page. The same page says visualizations do not render in the Codex CLI or the IDE extension. On the web they work for supported accounts in Chat and ChatGPT Work.
Which model you get depends on your plan. GPT-6 Sol runs the feature for Plus, Pro, Business and Enterprise, and GPT-6 Luna runs it for Free and Go. OpenAI says both are tuned for everyday conversation. These are the same two models that reached ChatGPT Work, Codex and the API on September 22, at half the GPT-5.6 price.
The release covers the Chat tab only. OpenAI says the models behind the other two tabs stay as they are: “The models powering Work and Codex are not changing as part of this release.”
Up next
Haiku 5.5 cuts token prices 90% on prompts under 100k, Video
At around 75% cheaper to run than Haiku 4.5, Claude Haiku 5.5 still scores 39.2% on Terminal-Bench 4.0. Its predecessor scored 0.0% on the same test.
Related stories
- OpenAI speeds up GPT-6 models by about 50% in ChatGPT
- OpenAI’s half-price GPT-6 Sol and Luna aren’t a promo
- ChatGPT users in the EU report getting OpenAI’s dots
- US Pro users get ChatGPT Wallet for agent payments
- US online businesses can publish their own agent in ChatGPT
- OpenAI opt-out skips personal Daybreak Blue sessions
Comments
No comments yet. Be the first.
Join the conversation
Sign in to leave a comment. It appears after moderation.

