OpenAI Opens GPT-6 to All Users, IntelligentUI Turns Answers Into Interactive Apps
机器之心 · wechat · 2026-10-08
OpenAI is rolling out GPT-6 to all ChatGPT users worldwide, free and paid, replacing GPT-5.6 Sol and GPT-5.6 Luna in ChatGPT; users accessing via Codex and ChatGPT Work keep the earlier versions.
- The headline capability, IntelligentUI, lets GPT-6 decide how to combine text, images, charts, buttons and forms into an interactive answer interface. Product comparisons render side by side, complex concepts become interactive diagrams, and simple questions still get plain text. It can also build task-specific tools on the spot, such as a savings-growth calculator, a bill-splitting tool or a playable mini-game inside the conversation.
- Technically, OpenAI built a streaming-native component library and an interface compiler: the model calls components and decides their composition while the compiler renders progressively, so users don't wait for the full answer. Training evaluates generated interfaces for clarity, usefulness and completeness, teaching the model when to add interactivity and when text suffices.
- On speed, GPT-6 interleaves reasoning and answering, emitting interim conclusions without filler. Internally, GPT-6 ExtraHigh's time-to-first-token matches GPT-5.6 Medium while its overall score beats GPT-5.6 ExtraHigh; for web-search questions GPT-6 Instant starts answering 44% earlier on average and judges better when to search.
- Availability: rolling out today to Plus, Pro, Business and Enterprise, expanding to Free and Go tomorrow; paid tiers run on GPT-6 Sol, free and Go on GPT-6 Luna, and the update applies only to the ChatGPT chat experience. Hours before launch, Sam Altman posted a cryptic line about awe under the stars, read as a nod to expanding model capability against human unknowns.
Related event: OpenAI Launches GPT-6 with Intelligent UI for All ChatGPT Users(60 posts)→
More from Models
- UK Turing Institute sovereign AI model roadmap targets deployable models late 2028 onwards — bclavie · 2026-11-19
- Gemini release gaps halve each cycle: 28 to 15 to 6 days per new model — davidyin44 · 2026-11-19
- Surge AI benchmarks Gemini 4 Neon: tops Tuesday@Work at 71.6, trails on SWE — echen · 2026-11-19
- Google's rumored Neon and Carbon models reportedly match or beat Opus in early tests — almmaasoglu · 2026-11-19
- Musk says Grok 4.8 is in RL phase, built for 'usefulness max' over benchmarks — XFreeze · 2026-11-19
- Liquid AI's d1.2-ultra beats Google and Microsoft decision models at half the price — JosephJacks_ · 2026-11-19