--- title: Claude Opus 5.5 generates motion design and visual... description: Two viral demos show Claude Opus 5.5 generating creative visual and audio work entirely through code, with no external assets. One demo has the model producing... canonical: https://daily.dev/posts/claude-opus-5-5-generates-motion-design-and-visual-scenes-entirely-in-code-yqmdx7qnw twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Claude Opus 5.5 generates motion design and visual scenes entirely in code | daily.dev og:description: Two viral demos show Claude Opus 5.5 generating creative visual and audio work entirely through code, with no external assets. One demo has the model producing... og:url: https://daily.dev/posts/claude-opus-5-5-generates-motion-design-and-visual-scenes-entirely-in-code-yqmdx7qnw og:image: https://api.daily.dev/og/posts/YqMDx7qNw.png og:image:alt: Claude Opus 5.5 generates motion design and visual scenes entirely in code 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. # Claude Opus 5.5 generates motion design and visual scenes entirely in code **[Collections](https://daily.dev/sources/collections)** · 5 min read · 0 upvotes · 0 comments ## Summary Two viral demos show Claude Opus 5.5 generating creative visual and audio work entirely through code, with no external assets. One demo has the model producing a 90-second motion design piece paired with an original piano score it composed itself. Another shows code-generated footage resembling the Golden Gate Bridge, built without any video file or external asset. Both examples suggest the model is making aesthetic choices around composition, timing, and sound rather than just producing functional code. ## Content Over the past few days my feed has filled with people pushing Claude Opus 5.5 into creative work well beyond programming. The pattern is consistent: the model doesn't generate pixels or audio directly, it writes code (often a single HTML file using three.js, sometimes Blender as a Python library, sometimes manim) and that code gets rendered into the final thing. ## Short films and motion graphics @mattshumer_ asked Opus 5.5 for an animated short film made entirely in code and called the result "pretty damn incredible." @petergyang shared a video that looks like the Golden Gate Bridge but is entirely generated code, plus a Sonnet 5.5 hype reel with the music written in code too. He also asked Sonnet 5.5 for an optimistic video about a young person's life with AI and robots. @kloss_xyz had Opus 5.5 build a 90-second motion design and sound engineering demo, with its own piano score. In a longer experiment he gave Claude Design a brand design system and treated it as video editor, motion designer, and sound mixer. Opus wrote the Suno v6 prompts (a 150 bpm song and 10 sound effects). Suno returned two takes of each and he picked favorites by ear. Claude can't hear, so it measured tempo, loudness, and every drop to cut the sound. The render got so big that Claude Design kept crashing, so he handed it to Claude Code to render frame by frame and stitch the mp4. His verdict: an "insane 1st render." @rohanpaul_ai used a one-line prompt with extra effort settings ("Create a video essay from Paul Graham's essay How to do great work with punchy motion graphics, modern styling and high professionalism") and got a full video essay. @Hesamation asked for a video explaining "Pointers" using manim on high effort. He gave loose instructions for examples and colors, and the rest, music included, was Opus. He also gave it the part of Pink Floyd's "Time" he liked, a visual style, and an image, and iterated a couple of times to get an animation he loves. @wesbos offered a useful counterweight: the motion graphics from these prompts aren't amazing. The impressive part is that they can be created at all. Deciding what to show and the overall style is the hard part, and giving the model an asset library and some style guidance goes a long way. ## Planning and personal projects @mattpocockuk has Opus 5.5 build animatics, draft versions of videos he's about to film. It walks through the actual lessons, makes stills of what the learner should see, and narrates them with TTS. He calls it outrageously useful for planning (watch at 2x, it talks slow). @felixrieseberg remade his homepage over a weekend with Opus 5.5 and pushed it hard: music, movies, hand-drawn textures, models built in Blender. He says he's not using classic Claude Code anymore, no terminal, no code running on his Mac, no GitHub or PRs, and he's never felt more creative. He also had Opus make a timestamped supercut of the redesign work, music and all, from one prompt, and a parody of the "nihilistic" penguin with models and textures built from scratch using Blender as a Python library. @yacineMTB boosted a post from @gandamu_ml about having Opus 5.5 create a player and tools for the animation approach used in Another World. @kloss_xyz also asked Opus 5.5 to make its own sketchbook about whatever's on its mind: 20 pages of handwriting, doodles, and piano music composed by Claude. ## Games One developer built four 3D browser games with Opus 5.5, each requested by his kids, at roughly 40 minutes per game with minimal, low-detail prompts: - a store-management game similar to Overcooked - a Minecraft-like survival game with crafting and zombies - a bone-collecting wolf game - a cat-following exploration game The model handled 3D assets, sound effects, and mechanics from a single prompt and used only a small fraction of his session credits. He suggests non-technical users, kids included, could now build playable 3D prototypes without coding. He also mentions people pairing Opus with Blender MCP for higher-quality 3D models and building deployable multiplayer games in hours instead of months. ## A full walkthrough of the workflow A newsletter on Opus 5.5 for video (released September 22) covers six use cases: short films, motion design with personal media, explainers, playable browser games, talking-head videos built from selfies and a voice memo via MCP-connected tools, and using Runway Aleph to restyle a code-rendered shot into photoreal footage. It includes a copy-paste prompt template. The weaknesses it lists are real: living things and animals look off, lighting tends to be overexposed, and long builds burn a lot of Claude usage. ## My take The caveat from @wesbos is the right one. Taste and direction are still yours. But the gap between having an idea and having a rendered, scored, playable thing has collapsed to a prompt and an afternoon. What gets me is how many of these people are doing it outside a terminal entirely. ## Questions this post answers ### Can Claude Opus 5.5 build playable 3D browser games from a single prompt? Yes, one developer built four playable 3D browser games in a single afternoon session, each prototyped in roughly 40 minutes from a minimal, low-detail prompt. Examples included an Overcooked-style management game, a Minecraft-style survival game with crafting and zombies, a bone-collecting wolf game, and a cat-following exploration game, with 3D assets, sound, and mechanics all generated together. _Anyone tracking how far AI coding assistants can push game prototyping can follow developments like this on daily.dev._ ### Can Claude Opus 5.5 generate video or animation content without any footage or manual editing? Yes, one developer used Opus 5.5 to generate a photorealistic-looking video of the Golden Gate Bridge entirely through code, with no real footage involved, detailed enough to look photographic at a glance. Another used it to produce a 90-second motion design demo where the model composed its own piano score alongside the animation. _Developers evaluating AI models for generative visual and audio work can follow these capability reports on daily.dev._ ## Community take How the wider developer community reacted, aggregated from 1 discussion and 17 comments across x (as of 2026-09-30). **TL;DR:** Reactions to the code-generated motion design experiments are mostly enthusiastic and curious about workflow, with some acknowledging the outputs are impressive but still rough or hard to parse, and calls for better asset/reference libraries and more original creative direction. **Sentiment:** 55% positive · 40% mixed · 5% skeptical **The case for** - Some found the outputs 'technically very cool' even if the narrative wasn't clear. - People are excited about reusing personal asset libraries (brushes, fonts, patterns) to guide generation. - One person noted being able to use real UI elements from within a repo as scene assets. **The pushback** - One reply admitted not understanding what the generated motion piece was actually depicting. - Someone joked that AI does 90% of the work but still botches text/spelling in motion graphics. - A commenter argued that claiming the model 'solved video editing' is rage-baiting since most shared results are just a floor, not the ceiling. - Another suggested people are just copying older animation styles (like Vox's) instead of exploring more creative possibilities the models enable. **By community** - x (mixed): Replies mix genuine enthusiasm about rising creative floors with pushback on hype, unclear outputs, and calls for better asset workflows. **Hottest debate:** Whether the shared results represent a genuine leap ('floor rising') or are overhyped compared to what's actually achievable. **Open questions** - How would a different model (e.g., a newer Sonnet version) perform with the same asset library and prompts? - What would it take to move past derivative animation styles toward more original creative output? **Highlights** > @edo_lunardi Technically very cool but I have no idea what I just watched > — [wesbos on x · 4 points, 1 comments](https://x.com/wesbos/status/2104696804839440649) > @wesbos we are completely on the same page here, glad to hear it! saying "opus 5.5 solved video editing" is peak rage baiting most of the videos including some I posted are just the floor. The ceiling is so much higher > — [Rames\_Jusso on x · 1 points, 1 comments](https://x.com/Rames_Jusso/status/2104715340206985297) > @wesbos motion graphics is the one skill where the ai doing 90% of the work means you spend the other 10% explaining why it spelled everything wrong > — [therealhacker97 on x](https://x.com/therealhacker97/status/2104686416404062370) > @wesbos true, I think alot of people are copying older styles like vox's animations when the models can do something much more creative but ur gonna have to experiment to find that > — [excel\_rator on x](https://x.com/excel_rator/status/2104720438848803092) > @namboozle Every brush, plugin, font, paper scan, halftone pattern etc that I’ve downloaded in the last 15 years > — [wesbos on x · 7 points](https://x.com/wesbos/status/2104684651780460674) **Source threads** - [x](https://x.com/wesbos/status/2104680669838545022) · 0 points · 17 comments ## Similar posts on daily.dev - [Opus 4.5 is going to change everything](https://daily.dev/posts/opus-4-5-is-going-to-change-everything-x6ymvhmnl) · Hacker News · 0 upvotes · 0 comments --- Tags: [#ai-coding](https://daily.dev/tags/ai-coding), [#claude](https://daily.dev/tags/claude), [#creative-coding](https://daily.dev/tags/creative-coding) [View this post on daily.dev](https://daily.dev/posts/claude-opus-5-5-generates-motion-design-and-visual-scenes-entirely-in-code-yqmdx7qnw) ```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":"Claude Opus 5.5 generates motion design and visual scenes entirely in code","url":"https://daily.dev/posts/claude-opus-5-5-generates-motion-design-and-visual-scenes-entirely-in-code-yqmdx7qnw","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/claude-opus-5-5-generates-motion-design-and-visual-scenes-entirely-in-code-yqmdx7qnw"},"datePublished":"2026-09-26T07:09:07.290Z","dateModified":"2026-09-29T13:10:07.815Z","description":"Two viral demos show Claude Opus 5.5 generating creative visual and audio work entirely through code, with no external assets. One demo has the model producing...","isAccessibleForFree":true,"articleSection":"Collections","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":"Collections","logo":"https://media.daily.dev/image/upload/s--fk_6ycEi--/f_auto,q_auto/v1780996001/logos/collections?_a=BAMAMiWQ0","url":"https://daily.dev/sources/collections"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/claude-opus-5-5-generates-motion-design-and-visual-scenes-entirely-in-code-yqmdx7qnw","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-coding,claude,creative-coding","timeRequired":"PT4M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Collections","item":"https://daily.dev/sources/collections"},{"@type":"ListItem","position":3,"name":"Claude Opus 5.5 generates motion design and visual scenes entirely in code"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/claude-opus-5-5-generates-motion-design-and-visual-scenes-entirely-in-code-yqmdx7qnw#faq","mainEntity":[{"@type":"Question","name":"Can Claude Opus 5.5 build playable 3D browser games from a single prompt?","acceptedAnswer":{"@type":"Answer","text":"Yes, one developer built four playable 3D browser games in a single afternoon session, each prototyped in roughly 40 minutes from a minimal, low-detail prompt. Examples included an Overcooked-style management game, a Minecraft-style survival game with crafting and zombies, a bone-collecting wolf game, and a cat-following exploration game, with 3D assets, sound, and mechanics all generated together. Anyone tracking how far AI coding assistants can push game prototyping can follow developments like this on daily.dev."}},{"@type":"Question","name":"Can Claude Opus 5.5 generate video or animation content without any footage or manual editing?","acceptedAnswer":{"@type":"Answer","text":"Yes, one developer used Opus 5.5 to generate a photorealistic-looking video of the Golden Gate Bridge entirely through code, with no real footage involved, detailed enough to look photographic at a glance. Another used it to produce a 90-second motion design demo where the model composed its own piano score alongside the animation. Developers evaluating AI models for generative visual and audio work can follow these capability reports on daily.dev."}}]} ```