--- title: Runway's Solaris generates UI frame by frame, no code... description: Runway launched Solaris, an 'Interface World Models' system that generates software UI frame by frame in real time based on user input, skipping code entirely.... canonical: https://daily.dev/posts/runway-s-solaris-generates-ui-frame-by-frame-no-code-required-s4pwdttql twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Runway's Solaris generates UI frame by frame, no code required | daily.dev og:description: Runway launched Solaris, an 'Interface World Models' system that generates software UI frame by frame in real time based on user input, skipping code entirely.... og:url: https://daily.dev/posts/runway-s-solaris-generates-ui-frame-by-frame-no-code-required-s4pwdttql og:image: https://api.daily.dev/og/posts/s4pwDTtql.png og:image:alt: Runway's Solaris generates UI frame by frame, no code required og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Runway's Solaris generates UI frame by frame, no code required **[Trends](https://daily.dev/sources/trends)** · 2 min read · 1 upvotes · 0 comments ## Summary Runway launched Solaris, an 'Interface World Models' system that generates software UI frame by frame in real time based on user input, skipping code entirely. It combines Runway's Gen-4.5 video model for visuals with a language model for reasoning. Runway's internal evals claim it beats screenshot-reconstruction models on fidelity and won 61-71% of comparisons against a Claude Opus 5-coded interface. Industry figures like Guillermo Rauch and Thomas Wolf weighed in on the long-term implications, with Wolf noting the tradeoff between predictability and adaptability in non-deterministic UI generation. Early access is open now ahead of a public launch. ## Content Runway just shipped something that makes the design-to-code pipeline look like a relic. Solaris, its new "Interface World Models" system, renders software interfaces in real time, frame by frame, directly from user input. No pre-coded layouts. No HTML. You click or drag, the model generates what happens next. The underlying stack is Runway's Gen-4.5 video model handling visuals, paired with a language model for reasoning. The pitch is that you skip the entire translation step between design intent and working interface. Runway's own evaluations claim Solaris preserved visual fidelity better than multimodal models reconstructing interfaces from screenshots, and that users preferred it over a Claude Opus 5-driven coded interface in 61-71% of comparisons. Those are internal numbers, so take them with appropriate skepticism, but the demos are doing real work here. The reaction from people who saw the clips was immediate. Omar Sarsur called it "nuts" and started listing what you could build: simulators, games, anything where the interface itself is part of the experience rather than a wrapper around it. Vercel's Guillermo Rauch framed it as part of a broader trend he's been watching: as models get more capable, UI gets more generative and dynamic. The more interesting take came from Hugging Face's Thomas Wolf, who sketched out where this goes long-term: ultrafast diffusion models handling interfaces, ultrafast LLMs predicting next application states. His framing was honest about the tradeoff - "what we'll lose in predictability we'll gain 100 folds in adaptability." That's the real question nobody's fully answered yet. Generative UI is responsive and flexible, but it's also non-deterministic. A button that renders differently each time is either a feature or a nightmare depending on your use case. Runway is opening early access now while working toward a public launch. The category name - Interface World Models - is clearly a positioning move, borrowing the "world model" framing from robotics and game AI. Whether that framing holds up as the technology matures is worth watching. ## Questions this post answers ### What is Runway Solaris and how does it generate user interfaces? Solaris is Runway's 'Interface World Models' system that renders software interfaces in real time, frame by frame, directly from user input like clicks or drags, with no pre-coded layouts or HTML involved. It combines Runway's Gen-4.5 video model for handling visuals with a separate language model for reasoning, skipping the translation step between design intent and working code. _Developers weighing generative UI against traditional code pipelines can follow tool comparisons like this on daily.dev._ ### How does Runway Solaris compare to a Claude Opus 5 coded interface in user preference tests? In Runway's own internal evaluations, users preferred Solaris over a Claude Opus 5-driven coded interface in 61-71% of comparisons, and Solaris preserved visual fidelity better than multimodal models reconstructing interfaces from screenshots. These are internal, self-reported numbers, so they warrant skepticism rather than being treated as independently verified benchmarks. _Anyone deciding between generative and coded UI approaches can track independent benchmarks on daily.dev._ ### What is the main tradeoff of using generative, non-deterministic UI models instead of coded interfaces? Generative UI systems like Solaris trade predictability for adaptability: a button or layout can render differently each time, which is either a feature or a nightmare depending on the use case. Hugging Face's Thomas Wolf framed this as losing predictability while gaining adaptability roughly a hundredfold, describing a future where ultrafast diffusion models render interfaces and ultrafast LLMs predict the next application state. _Teams debating deterministic versus generative interface architectures can follow this tradeoff unfold on daily.dev._ ## Similar posts on daily.dev - [Runway releases its first world model, adds native audio to latest video model](https://daily.dev/posts/runway-releases-its-first-world-model-adds-native-audio-to-latest-video-model-8471ciojo) · TechCrunch · 1 upvotes · 0 comments - [Runway says its new text-to-video AI generator has ‘unprecedented’ accuracy](https://daily.dev/posts/runway-says-its-new-text-to-video-ai-generator-has-unprecedented-accuracy-acxpsywsu) · The Verge · 0 upvotes · 0 comments --- Tags: [#ai-coding](https://daily.dev/tags/ai-coding), [#diffusion-models](https://daily.dev/tags/diffusion-models) [View this post on daily.dev](https://daily.dev/posts/runway-s-solaris-generates-ui-frame-by-frame-no-code-required-s4pwdttql) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Runway's Solaris generates UI frame by frame, no code required","url":"https://daily.dev/posts/runway-s-solaris-generates-ui-frame-by-frame-no-code-required-s4pwdttql","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/runway-s-solaris-generates-ui-frame-by-frame-no-code-required-s4pwdttql"},"datePublished":"2026-09-01T20:22:33.761Z","dateModified":"2026-09-01T20:23:18.146Z","description":"Runway launched Solaris, an 'Interface World Models' system that generates software UI frame by frame in real time based on user input, skipping code entirely....","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/242798ee82c32f00d44df767d42abc19?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/242798ee82c32f00d44df767d42abc19?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Trends","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Trends","logo":"https://media.daily.dev/image/upload/s--ZfSp3asX--/f_auto,q_auto/v1780996004/logos/trends?_a=BAMAMiWQ0","url":"https://daily.dev/sources/trends"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/runway-s-solaris-generates-ui-frame-by-frame-no-code-required-s4pwdttql","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-coding,diffusion-models","timeRequired":"PT2M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Trends","item":"https://daily.dev/sources/trends"},{"@type":"ListItem","position":3,"name":"Runway's Solaris generates UI frame by frame, no code required"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/runway-s-solaris-generates-ui-frame-by-frame-no-code-required-s4pwdttql#faq","mainEntity":[{"@type":"Question","name":"What is Runway Solaris and how does it generate user interfaces?","acceptedAnswer":{"@type":"Answer","text":"Solaris is Runway's 'Interface World Models' system that renders software interfaces in real time, frame by frame, directly from user input like clicks or drags, with no pre-coded layouts or HTML involved. It combines Runway's Gen-4.5 video model for handling visuals with a separate language model for reasoning, skipping the translation step between design intent and working code. Developers weighing generative UI against traditional code pipelines can follow tool comparisons like this on daily.dev."}},{"@type":"Question","name":"How does Runway Solaris compare to a Claude Opus 5 coded interface in user preference tests?","acceptedAnswer":{"@type":"Answer","text":"In Runway's own internal evaluations, users preferred Solaris over a Claude Opus 5-driven coded interface in 61-71% of comparisons, and Solaris preserved visual fidelity better than multimodal models reconstructing interfaces from screenshots. These are internal, self-reported numbers, so they warrant skepticism rather than being treated as independently verified benchmarks. Anyone deciding between generative and coded UI approaches can track independent benchmarks on daily.dev."}},{"@type":"Question","name":"What is the main tradeoff of using generative, non-deterministic UI models instead of coded interfaces?","acceptedAnswer":{"@type":"Answer","text":"Generative UI systems like Solaris trade predictability for adaptability: a button or layout can render differently each time, which is either a feature or a nightmare depending on the use case. Hugging Face's Thomas Wolf framed this as losing predictability while gaining adaptability roughly a hundredfold, describing a future where ultrafast diffusion models render interfaces and ultrafast LLMs predict the next application state. Teams debating deterministic versus generative interface architectures can follow this tradeoff unfold on daily.dev."}}]} ```