Video wird geladen...

Video konnte nicht geladen werden

Beim Laden dieses Videos ist ein Problem aufgetreten. Dies könnte an einem vorübergehenden Netzwerkproblem liegen oder das Video ist möglicherweise nicht verfügbar.

A visual comparison of HTML-in-Canvas vs rasterizing DOM to a canvas texture (side-by-side below, compare average capture times) Existing approaches to pull web content into canvas are slow and lossy: - Animations glitch, or don't play at all - CSS replication is imprecise - Clogs up the main thread,... show more

Robin

3,988 subscribers

46,497 Aufrufe • vor 3 Monaten •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

HTML can't run Doom But HTML can now run *inside* Doom! Thanks to HTML-in-Canvas! Fully accessible DOM elements drawn into Doom's own wall textures. This demo uses a WebAssembly port (jacobenget/doom.wasm) of the original C source. See it in action at - Requires Chrome with chrome://flags/#canvas-draw-element enabled. On every Doom tick, HTML-in-Canvas's drawElementImage() captures the DOM form (child of ) to an offscreen buffer, it's quantised to Doom's 256-colour palette, and the bytes are written straight into the cached patch for the appropriate texture in Doom. Doom's own software renderer then paints it onto E1M1 as a wall texture - which means perspective warp, sector lighting, and even the damage-red tint all get included and applied to the HTML form elements. It's just a texture to Doom.

HTML can't run Doom But HTML can now run inside Doom! Thanks to HTML-in-Canvas! Fully accessible DOM elements drawn into Doom's own wall textures. This demo uses a WebAssembly port (jacobenget/doom.wasm) of the original C source. See it in action at - Requires Chrome with chrome://flags/#canvas-draw-element enabled. On every Doom tick, HTML-in-Canvas's drawElementImage() captures the DOM form (child of ) to an offscreen buffer, it's quantised to Doom's 256-colour palette, and the bytes are written straight into the cached patch for the appropriate texture in Doom. Doom's own software renderer then paints it onto E1M1 as a wall texture - which means perspective warp, sector lighting, and even the damage-red tint all get included and applied to the HTML form elements. It's just a texture to Doom.

Robin

114,553 Aufrufe • vor 3 Monaten

New short course: Collaborative Writing and Coding with OpenAI Canvas! Explore new ways to write and code with OpenAI Canvas, a user-friendly interface that allows you to brainstorm, draft, and refine text and code in collaboration with ChatGPT. In the short course, created with OpenAI, and taught by , a research lead at OpenAI, you’ll learn to use Canvas to enhance your workflows. Canvas lets you go beyond simple chat interactions. It provides a side-by-side workspace where you and ChatGPT can edit and refine text or code collaboratively. This makes brainstorming, drafting, and iterating as you write feel more natural and effective. As the first major update to ChatGPT’s visual interface since its launch in 2022, Canvas gives a new, innovative approach to collaboration with AI. For instance, after writing the first version of your code, Canvas can review it and give suggestions for improvement. It can also help with debugging by adding logging, identifying problems to fix, and writing comments. In addition, you'll also learn what it takes to train the model for an interface like Canvas. In this video-only short course, you’ll: - Learn how to ask for in-line feedback and control the iteration of your work by directly editing selected areas of your text or code from the model’s output. - Learn how to access quick automation tools in a shortcut menu that allows you to modify your writing tone and length, enhance your code, and restore previous versions of your work. - Learn how to use Canvas as a research assistant tool with an example of asking the model to reason through the screenshot of a plot to write a research report, in which you can ask questions within the created report. - Ask the model to write Python code to replicate the graph seen on a screenshot image. - Go behind the scenes of how you can create a video game, such as Space Battleship, from scratch, edit it, and display it in one self-contained HTML file. - Get a real-world application example of creating a SQL database from the image of its architecture. - Understand the model training and design processes that power Canvas! Please sign up here:

New short course: Collaborative Writing and Coding with OpenAI Canvas! Explore new ways to write and code with OpenAI Canvas, a user-friendly interface that allows you to brainstorm, draft, and refine text and code in collaboration with ChatGPT. In the short course, created with OpenAI, and taught by , a research lead at OpenAI, you’ll learn to use Canvas to enhance your workflows. Canvas lets you go beyond simple chat interactions. It provides a side-by-side workspace where you and ChatGPT can edit and refine text or code collaboratively. This makes brainstorming, drafting, and iterating as you write feel more natural and effective. As the first major update to ChatGPT’s visual interface since its launch in 2022, Canvas gives a new, innovative approach to collaboration with AI. For instance, after writing the first version of your code, Canvas can review it and give suggestions for improvement. It can also help with debugging by adding logging, identifying problems to fix, and writing comments. In addition, you'll also learn what it takes to train the model for an interface like Canvas. In this video-only short course, you’ll: - Learn how to ask for in-line feedback and control the iteration of your work by directly editing selected areas of your text or code from the model’s output. - Learn how to access quick automation tools in a shortcut menu that allows you to modify your writing tone and length, enhance your code, and restore previous versions of your work. - Learn how to use Canvas as a research assistant tool with an example of asking the model to reason through the screenshot of a plot to write a research report, in which you can ask questions within the created report. - Ask the model to write Python code to replicate the graph seen on a screenshot image. - Go behind the scenes of how you can create a video game, such as Space Battleship, from scratch, edit it, and display it in one self-contained HTML file. - Get a real-world application example of creating a SQL database from the image of its architecture. - Understand the model training and design processes that power Canvas! Please sign up here:

Andrew Ng

128,180 Aufrufe • vor 1 Jahr

Here is a quick walkthrough of everything new in Stitch: 🌊 Streaming: Stitch now streams its work straight to the canvas. You can see Stitch working and start steering before the final product is done! 📄 Existing Designs: Stitch can help you start with your design by creating a DESIGN.md from your codebase, .fig files, or a live website ⏩ In-place AI Edits: Stitch can quickly edit specific parts of your screens just by asking or pointing and clicking 🎥 Motion & HTML canvas: Javascript, shaders and more now render natively on an HTML canvas so you can immediately see animation, hover states and effects 🚀 Portability and Integrations: We've added a big list of import and export additions to fit right into your workflow, including import your .fig files and one-click export to Netlify, Loveable and Bolt 🌟 Bonus: New skills to IMPORT your screens directly into Stitch via MCP and sync changes back to your codebase Walkthrough video👇 Details for each improvement in 🧵

Here is a quick walkthrough of everything new in Stitch: 🌊 Streaming: Stitch now streams its work straight to the canvas. You can see Stitch working and start steering before the final product is done! 📄 Existing Designs: Stitch can help you start with your design by creating a DESIGN.md from your codebase, .fig files, or a live website ⏩ In-place AI Edits: Stitch can quickly edit specific parts of your screens just by asking or pointing and clicking 🎥 Motion & HTML canvas: Javascript, shaders and more now render natively on an HTML canvas so you can immediately see animation, hover states and effects 🚀 Portability and Integrations: We've added a big list of import and export additions to fit right into your workflow, including import your .fig files and one-click export to Netlify, Loveable and Bolt 🌟 Bonus: New skills to IMPORT your screens directly into Stitch via MCP and sync changes back to your codebase Walkthrough video👇 Details for each improvement in 🧵

Stitch by Google

21,597 Aufrufe • vor 1 Monat

Introducing Slashspace✨ A canvas first agentic harness for power AI users. We are trying to replace all UIs with one interface for the AI era. For the past 18 months i've been building this and it's finally taking shape towards being the ultimate AI interface for a power AI user. We iterated on the canvas experience many times. Started as RabbitHoles AI with node based conversations. Slashspace is now evolving towards canvas-as-context app - The nodes are all agentic, and you don't need to draw connections and do the manual work. Today we're launching on Product Hunt 😸, and your support will mean a lot! As a solo bootstrapped founder, I've invested a lot into this product, so please show some love! 👀 Some feature highlights • Canvas with multi model, multi modal chats. • Local storage with all canvases stored in your folder system • Agentic nodes that can see all the context on the canvas. • Connect to Cursor sdk and use the cursor's harness locally, but chat on the canvas • Access to all the Composio supported tools on the canvas. s/o KP for hunting our PH launch. s/o to Tibo, Tony Dinh, David & HRG gang for forcing the clarity out of me on repositioning the product

Introducing Slashspace✨ A canvas first agentic harness for power AI users. We are trying to replace all UIs with one interface for the AI era. For the past 18 months i've been building this and it's finally taking shape towards being the ultimate AI interface for a power AI user. We iterated on the canvas experience many times. Started as RabbitHoles AI with node based conversations. Slashspace is now evolving towards canvas-as-context app - The nodes are all agentic, and you don't need to draw connections and do the manual work. Today we're launching on Product Hunt 😸, and your support will mean a lot! As a solo bootstrapped founder, I've invested a lot into this product, so please show some love! 👀 Some feature highlights • Canvas with multi model, multi modal chats. • Local storage with all canvases stored in your folder system • Agentic nodes that can see all the context on the canvas. • Connect to Cursor sdk and use the cursor's harness locally, but chat on the canvas • Access to all the Composio supported tools on the canvas. s/o KP for hunting our PH launch. s/o to Tibo, Tony Dinh, David & HRG gang for forcing the clarity out of me on repositioning the product

Praneeth Pike

44,277 Aufrufe • vor 1 Monat

gm gm. today is a special day as i have accomplished something i never thought i would be able to do. i started apes on canvas almost three years ago. i've done over a hundred oil on canvases, from projects such as The Ape Society, Pudgy Penguins, DeGods, among many others. but today, thanks to the painting comissioned by ESCO 🪏, i have done my first ever Yuga Labs MAYC. the paintings are oil on canvas and are framed in a baroque style frame, with about a tenth of an ounce of gold laid into the frames in the form of 24k gold leaf. this particular painting also included a first ever gold leaf applied directly to the canvas to highlight the gold coins dripping out of the ape's eyes. here is the final result, i hope you all like it:

gm gm. today is a special day as i have accomplished something i never thought i would be able to do. i started apes on canvas almost three years ago. i've done over a hundred oil on canvases, from projects such as The Ape Society, Pudgy Penguins, DeGods, among many others. but today, thanks to the painting comissioned by ESCO 🪏, i have done my first ever Yuga Labs MAYC. the paintings are oil on canvas and are framed in a baroque style frame, with about a tenth of an ounce of gold laid into the frames in the form of 24k gold leaf. this particular painting also included a first ever gold leaf applied directly to the canvas to highlight the gold coins dripping out of the ape's eyes. here is the final result, i hope you all like it:

CAЯVEЯ de Medici 🎩🐸

37,113 Aufrufe • vor 1 Jahr

Forget trying to design everything in Figma. You can build your own design tools with AI. Introducing Toolcraft - a starter kit and UI library for building beautiful creative apps. It's free and available for you today. npx @pixel-point/toolcraft create Use it to build apps to stylize images, create WebGL shaders, Three.js scenes, animations, photo editing tools, or whatever else needs a canvas and controls. Toolcraft gives you the full architecture in a single solution. It comes with: - plenty of UI components: sliders, pickers, timelines, curves, and more - built-in canvas, export, and toolbar setup - AI instructions to help the agent deliver a great result to you With Toolcraft, you don’t need an extensive prompt. Just explain what type of visuals you want to build, attach references, and wait. It’s been a game changer for our design process. Watch the full video on our YouTube channel with all the details and examples.

Forget trying to design everything in Figma. You can build your own design tools with AI. Introducing Toolcraft - a starter kit and UI library for building beautiful creative apps. It's free and available for you today. npx @pixel-point/toolcraft create Use it to build apps to stylize images, create WebGL shaders, Three.js scenes, animations, photo editing tools, or whatever else needs a canvas and controls. Toolcraft gives you the full architecture in a single solution. It comes with: - plenty of UI components: sliders, pickers, timelines, curves, and more - built-in canvas, export, and toolbar setup - AI instructions to help the agent deliver a great result to you With Toolcraft, you don’t need an extensive prompt. Just explain what type of visuals you want to build, attach references, and wait. It’s been a game changer for our design process. Watch the full video on our YouTube channel with all the details and examples.

Alex Barashkov

260,884 Aufrufe • vor 15 Tagen

This might be the biggest evolution to vibe coding You’re not limited to basic websites anymore but you can generate: - Full stack web and mobile apps - Pro animations to promote them - Slides, documents or data apps Everything at the same place and in a single infinite canvas using Replit Agent 4 and most importantly… IN PARALLEL! So you don't have to wait until one agent finishes a minor modification to continue working on another part of your project. Future really belongs to builders. Technical barriers don’t exist anymore, that’s exciting! (You can give it a try with the link below)

This might be the biggest evolution to vibe coding You’re not limited to basic websites anymore but you can generate: - Full stack web and mobile apps - Pro animations to promote them - Slides, documents or data apps Everything at the same place and in a single infinite canvas using Replit Agent 4 and most importantly… IN PARALLEL! So you don't have to wait until one agent finishes a minor modification to continue working on another part of your project. Future really belongs to builders. Technical barriers don’t exist anymore, that’s exciting! (You can give it a try with the link below)

Paul Couvert

11,264 Aufrufe • vor 4 Monaten

The web was never meant to be flattened into text. Yet most web RAG systems start by parsing HTML --- a complex and lossy process. 🔥 Introducing PixelRAG: the first RAG system that retrieves and reads 30M+ web pages as pixels. Instead of extracting text, PixelRAG retrieves screenshots and lets a VLM read them directly. PixelRAG not only preserves visual information, but also outperforms text-based RAG on text-only QA benchmarks by +18.1%. Why? (1) HTML-to-text conversion often discards layout, structure, tables, and other useful signals. (2) We continued pretraining a VLM on web page screenshots and turned it into a surprisingly strong visual retriever. (3) Recent VLMs are remarkably good at understanding web pages, often with better accuracy and token efficiency than text-only pipelines. Takeaway: HTML parsing may be one of the biggest self-inflicted bottlenecks in web RAG. Demo below 👇 Code: Paper: Playground:

The web was never meant to be flattened into text. Yet most web RAG systems start by parsing HTML --- a complex and lossy process. 🔥 Introducing PixelRAG: the first RAG system that retrieves and reads 30M+ web pages as pixels. Instead of extracting text, PixelRAG retrieves screenshots and lets a VLM read them directly. PixelRAG not only preserves visual information, but also outperforms text-based RAG on text-only QA benchmarks by +18.1%. Why? (1) HTML-to-text conversion often discards layout, structure, tables, and other useful signals. (2) We continued pretraining a VLM on web page screenshots and turned it into a surprisingly strong visual retriever. (3) Recent VLMs are remarkably good at understanding web pages, often with better accuracy and token efficiency than text-only pipelines. Takeaway: HTML parsing may be one of the biggest self-inflicted bottlenecks in web RAG. Demo below 👇 Code: Paper: Playground:

Yichuan Wang

89,470 Aufrufe • vor 1 Monat

Being able to move fluidly between Figma Make and the canvas is a pretty big deal Recent example 👇 1️⃣ Started exploring a minified recording toolbar directly on the canvas 2️⃣ Once I had some rough shapes I moved it into Make, imported my style guide, and started bringing it to life with expandable controls, hover states, and tooltips for keyboard shortcuts 3️⃣ A few prompts later, I had something interesting… but the layout wasn’t quite right. 4️⃣ So I jumped back into the canvas and made ~8 quick variations with a bigger video container. 5️⃣ Once I got somewhere interesting I brought that back into Make. All I had to do was copy and paste the frames for my default and hover state of the component and it pretty much nailed the core layout 6️⃣ Then I layered on the timer and made it interactive. But once I got it back into Make I thought "hmm... what if the toolbar layout was based on which quadrant of the screen the component was in 🤔" I was able to one shot this interaction and after playing with it a bunch I kinda love it? 7️⃣ Sent it to the team in Inflight so everyone could play with the prototype. 8️⃣ Brought the up to date UI back into the canvas to polish some of the tooltip details Before this, it always felt like I had to choose between two very different workflows. Now I can bounce between them seamlessly and it’s genuinely changing how I explore ideas.

Being able to move fluidly between Figma Make and the canvas is a pretty big deal Recent example 👇 1️⃣ Started exploring a minified recording toolbar directly on the canvas 2️⃣ Once I had some rough shapes I moved it into Make, imported my style guide, and started bringing it to life with expandable controls, hover states, and tooltips for keyboard shortcuts 3️⃣ A few prompts later, I had something interesting… but the layout wasn’t quite right. 4️⃣ So I jumped back into the canvas and made ~8 quick variations with a bigger video container. 5️⃣ Once I got somewhere interesting I brought that back into Make. All I had to do was copy and paste the frames for my default and hover state of the component and it pretty much nailed the core layout 6️⃣ Then I layered on the timer and made it interactive. But once I got it back into Make I thought "hmm... what if the toolbar layout was based on which quadrant of the screen the component was in 🤔" I was able to one shot this interaction and after playing with it a bunch I kinda love it? 7️⃣ Sent it to the team in Inflight so everyone could play with the prototype. 8️⃣ Brought the up to date UI back into the canvas to polish some of the tooltip details Before this, it always felt like I had to choose between two very different workflows. Now I can bounce between them seamlessly and it’s genuinely changing how I explore ideas.

Ridd 🤿

27,665 Aufrufe • vor 8 Monaten

THIS GUY GOT SICK OF OPENING 30 TABS ON WIKIPEDIA SO HE VIBE CODED A VISUAL BROWSER FOR IT we all know the wikipedia rabbit hole. you click one article, then another link inside it, then another, then another. 30 tabs deep and you forgot where you started this guy built a canvas based browser where every article you click opens as a card on an infinite canvas. every link between articles is a visible connection you can: > look up any article > browse the home screen > draw connections > add post-it nodes > even save your boards locally instead of losing yourself in tabs you can see your entire research path laid out visually like a mind map click an article. the linked topics branch out around it. click one of those. it branches further. your entire rabbit hole is one visual map you can zoom in and out of someone said that this should exist for the entire internet the concept is called "canvas-like browsing" and its a great way to do research reading wikipedia has always been the best way to learn about anything and now you can actually see how everything connects

THIS GUY GOT SICK OF OPENING 30 TABS ON WIKIPEDIA SO HE VIBE CODED A VISUAL BROWSER FOR IT we all know the wikipedia rabbit hole. you click one article, then another link inside it, then another, then another. 30 tabs deep and you forgot where you started this guy built a canvas based browser where every article you click opens as a card on an infinite canvas. every link between articles is a visible connection you can: > look up any article > browse the home screen > draw connections > add post-it nodes > even save your boards locally instead of losing yourself in tabs you can see your entire research path laid out visually like a mind map click an article. the linked topics branch out around it. click one of those. it branches further. your entire rabbit hole is one visual map you can zoom in and out of someone said that this should exist for the entire internet the concept is called "canvas-like browsing" and its a great way to do research reading wikipedia has always been the best way to learn about anything and now you can actually see how everything connects

Om Patel

14,437 Aufrufe • vor 1 Monat

PROMPT DROP 🚨 for Creators & NFT Collectors. 🎨 Minting the Self: Genesis of the Living Canvas🪽 The prompt is optimised to inherit the style, colour palette, and personality of your character. 💀 Drop your creations below 🌐 Keen to see what you mint 👀 🚨PROMPT👇🏼 Use the provided PFP/NFT character as the primary reference. The character must remain fully recognizable and true to its original design—preserve exact facial features, proportions, textures, materials, color palette, accessories, and overall personality. All elements in the scene must adapt to and harmonize with the character’s unique visual style—whether that is cartoon, anime, voxel, hyper-realistic, or stylized 3D. Create a hyper-detailed fine art studio scene where the character stands before a massive canvas. The painting on the canvas is a fully completed, master-level version of the same character, rendered in a refined, elevated interpretation of its original style. The character’s signature identity feature (mask, face, helmet, eyes, or defining trait) must be 100% complete, sharp, and highly detailed—serving as the focal anchor of the composition. They hold a paintbrush dripping with material true to their world (oil paint, neon energy, pixels, liquid chrome, etc.), and the colors precisely match their original palette—visually linking the act of painting to the formation of their body. The studio environment subtly inherits the character’s aesthetic and personality: Color grading aligns with the character’s palette (dark, vibrant, pastel, neon, etc.) Props and materials reflect their universe (futuristic, street, fantasy, minimal, luxury) Lighting style matches their mood (soft cinematic, harsh contrast, neon glow, ethereal light) A large window casts natural or stylized light consistent with the character’s world, enhancing texture, depth, and atmosphere. Paint or material drips from both the canvas and the character, seamlessly blending creator and creation, reality and artwork. Style Output: ultra-detailed, cinematic composition, faithful to the character’s original art direction, enhanced with fine art realism and high-end 3D rendering, 8K resolution, museum-quality lighting, rich textures, depth of field, and tactile detail.

PROMPT DROP 🚨 for Creators & NFT Collectors. 🎨 Minting the Self: Genesis of the Living Canvas🪽 The prompt is optimised to inherit the style, colour palette, and personality of your character. 💀 Drop your creations below 🌐 Keen to see what you mint 👀 🚨PROMPT👇🏼 Use the provided PFP/NFT character as the primary reference. The character must remain fully recognizable and true to its original design—preserve exact facial features, proportions, textures, materials, color palette, accessories, and overall personality. All elements in the scene must adapt to and harmonize with the character’s unique visual style—whether that is cartoon, anime, voxel, hyper-realistic, or stylized 3D. Create a hyper-detailed fine art studio scene where the character stands before a massive canvas. The painting on the canvas is a fully completed, master-level version of the same character, rendered in a refined, elevated interpretation of its original style. The character’s signature identity feature (mask, face, helmet, eyes, or defining trait) must be 100% complete, sharp, and highly detailed—serving as the focal anchor of the composition. They hold a paintbrush dripping with material true to their world (oil paint, neon energy, pixels, liquid chrome, etc.), and the colors precisely match their original palette—visually linking the act of painting to the formation of their body. The studio environment subtly inherits the character’s aesthetic and personality: Color grading aligns with the character’s palette (dark, vibrant, pastel, neon, etc.) Props and materials reflect their universe (futuristic, street, fantasy, minimal, luxury) Lighting style matches their mood (soft cinematic, harsh contrast, neon glow, ethereal light) A large window casts natural or stylized light consistent with the character’s world, enhancing texture, depth, and atmosphere. Paint or material drips from both the canvas and the character, seamlessly blending creator and creation, reality and artwork. Style Output: ultra-detailed, cinematic composition, faithful to the character’s original art direction, enhanced with fine art realism and high-end 3D rendering, 8K resolution, museum-quality lighting, rich textures, depth of field, and tactile detail.

BiBi

11,129 Aufrufe • vor 3 Monaten

Kim namgil talking about canvas of blood and Bogum : in Korea, when people talk about historical figures, those who are evil or who seize power are usually portrayed as sharp looking, while the people trying to protect something are often cast with actors who have gentle, deer like eyes so in canvas of blood, even though it wasn’t intentional, i naturally gave off more of the image of someone trying to take something away, while Bogum played the prince trying to protect it, even just standing side by side, anyone can immediately tell the difference between the one trying to take and the one trying to protect, but on set, it was actually a little unclear between us, It became fun because you actually wonder who’s the one taking and who’s the one protecting, I think that’ll make the movie more enjoyable to watch Bogum always has this bright energy about him, which is really nice and because of that, there’s this anticipation about what kind of charm he’d show if he played someone more evil or morally neutral, so when people see this side of him this time, I think they’ll be really surprised HELLO? drop the movie already I really can’t wait 😭😭😭 #CanvasOfBlood

Kim namgil talking about canvas of blood and Bogum : in Korea, when people talk about historical figures, those who are evil or who seize power are usually portrayed as sharp looking, while the people trying to protect something are often cast with actors who have gentle, deer like eyes so in canvas of blood, even though it wasn’t intentional, i naturally gave off more of the image of someone trying to take something away, while Bogum played the prince trying to protect it, even just standing side by side, anyone can immediately tell the difference between the one trying to take and the one trying to protect, but on set, it was actually a little unclear between us, It became fun because you actually wonder who’s the one taking and who’s the one protecting, I think that’ll make the movie more enjoyable to watch Bogum always has this bright energy about him, which is really nice and because of that, there’s this anticipation about what kind of charm he’d show if he played someone more evil or morally neutral, so when people see this side of him this time, I think they’ll be really surprised HELLO? drop the movie already I really can’t wait 😭😭😭 #CanvasOfBlood

gummy 🇵🇸

14,709 Aufrufe • vor 1 Monat

i did it. i vibe-coded my own vibe-noding app⚡️ lately every ai tool has been announcing some kind of node canvas. they’re all impressive in their own way, but each one solves a different problem, with a different logic, in a different universe. some are too technical, some are too rigid, most feel like they were designed by a backend engineer. so i built the version i actually want to use. over the last few days i hacked this together with Google Gemini and Google Antigravity. the tool is absurdly capable. i didn’t touch a single line of code. the result is a fully working app with smooth ui animations, smart interactions, and a workflow that doesn’t fight me. my focus isn’t on the models. Nano Banana 2 Lite is already strong enough to generate half a universe on its own. and a year from now, we won’t care about model juggling anyway. we’ll have one model doing everything. the real problem is the workflow. the interface. the flow state. so i optimized for that. the whole point is to keep the technical junk out of the way during ideation and iteration. i want to try an idea, branch it, remix it, explore it, all on a single canvas. every block is editable. every idea is forkable. every path is visible. the design isn’t perfect yet. some of the ui patterns are familiar on purpose. good ideas deserve to be stolen. i rushed it. i let a bunch of great ui decisions die for speed. but it works. and honestly, it feels great to use. future of creating is more organic and more fluid. not just chat bubbles. not rigid feeds. but an intelligent canvas where ideas can grow, split, collapse, or evolve without friction. where the environment quietly adapts to your workflow instead of interrupting it. this is my first version. just for myself, for now. but it’s a start.

i did it. i vibe-coded my own vibe-noding app⚡️ lately every ai tool has been announcing some kind of node canvas. they’re all impressive in their own way, but each one solves a different problem, with a different logic, in a different universe. some are too technical, some are too rigid, most feel like they were designed by a backend engineer. so i built the version i actually want to use. over the last few days i hacked this together with Google Gemini and Google Antigravity. the tool is absurdly capable. i didn’t touch a single line of code. the result is a fully working app with smooth ui animations, smart interactions, and a workflow that doesn’t fight me. my focus isn’t on the models. Nano Banana 2 Lite is already strong enough to generate half a universe on its own. and a year from now, we won’t care about model juggling anyway. we’ll have one model doing everything. the real problem is the workflow. the interface. the flow state. so i optimized for that. the whole point is to keep the technical junk out of the way during ideation and iteration. i want to try an idea, branch it, remix it, explore it, all on a single canvas. every block is editable. every idea is forkable. every path is visible. the design isn’t perfect yet. some of the ui patterns are familiar on purpose. good ideas deserve to be stolen. i rushed it. i let a bunch of great ui decisions die for speed. but it works. and honestly, it feels great to use. future of creating is more organic and more fluid. not just chat bubbles. not rigid feeds. but an intelligent canvas where ideas can grow, split, collapse, or evolve without friction. where the environment quietly adapts to your workflow instead of interrupting it. this is my first version. just for myself, for now. but it’s a start.

Synthetic_soul

239,298 Aufrufe • vor 7 Monaten

I was laid off so I rebuilt their product but better (in 2 weeks from scratch) Open Canvas - a multiplayer site builder with an agent at the cursor. Most website builders ask you to fill in a template.I built one that lets you design - on a free-form canvas, with an AI agent at your cursor, where your whole team co-edits live and visitors see your publish update without refreshing. What makes it different: • Drag 14 design primitives anywhere — not into fixed slots • AI agent edits in plain English, you preview before it lands • Real-time co-editing (Yjs CRDT, Figma-style cursors) • Publish → every open visitor tab updates in <100ms Under the hood: one Cloudflare Worker, serverless Postgres, a first-class CMS, version history, a blocking accessibility audit, AI image gen, custom domains — all shipping as pure HTML, zero client-framework weight. It's MIT-licensed and built in the open

I was laid off so I rebuilt their product but better (in 2 weeks from scratch) Open Canvas - a multiplayer site builder with an agent at the cursor. Most website builders ask you to fill in a template.I built one that lets you design - on a free-form canvas, with an AI agent at your cursor, where your whole team co-edits live and visitors see your publish update without refreshing. What makes it different: • Drag 14 design primitives anywhere — not into fixed slots • AI agent edits in plain English, you preview before it lands • Real-time co-editing (Yjs CRDT, Figma-style cursors) • Publish → every open visitor tab updates in <100ms Under the hood: one Cloudflare Worker, serverless Postgres, a first-class CMS, version history, a blocking accessibility audit, AI image gen, custom domains — all shipping as pure HTML, zero client-framework weight. It's MIT-licensed and built in the open

Aayushman Singh

193,695 Aufrufe • vor 1 Monat

Gemini 2.5 Flash demolishes my Galton Board test, I could not get 4omini, 4o mini high, or 03 to produce this. I found that Gemini 2.5 Flash understands my intents almost instantly, code produced is tight and neat. The prompt is a merging of various steps. It took me 5 steps to achieve this in Gemini 2.5 Flash, I gave up on OpenAI models after about half an hour. My iterations are obviously not exact. But people can test with this one prompt for more objective comparison. Please try this prompt on your end to confirm: -------------------------------------------------- Create a self-contained HTML file for a Galton board simulation using client-side JavaScript and a 2D physics engine (like Matter.js, included via CDN). The simulation should be rendered on an HTML5 canvas and meet the following criteria: 1. **Single File:** All necessary HTML, CSS, and JavaScript code must be within this single `.html` file. 2. **Canvas Size:** The overall simulation area (canvas) should be reasonably sized to fit on a standard screen without requiring extensive scrolling or zooming (e.g., around 500x700 pixels). 3. **Physics:** Utilize a 2D rigid body physics engine for realistic ball-peg and ball-wall interactions. 4. **Obstacles (Pegs):** Create static, circular pegs arranged in full-width horizontal rows extending across the usable width of the board (not just a triangle). The pegs should be small enough and spaced appropriately for balls to navigate and bounce between them. 5. **Containment:** * Include static, sufficiently thick side walls and a ground at the bottom to contain the balls within the board. * Implement *physical* static dividers between the collection bins at the bottom. These dividers must be thick enough to prevent balls from passing through them, ensuring accurate accumulation in each bin. 6. **Ball Dropping:** Balls should be dropped from a controlled, narrow area near the horizontal center at the top of the board to ensure they enter the peg field consistently. 7. **Bins:** The collection area at the bottom should be divided into distinct bins by the physical dividers. The height of the bins should be sufficient to clearly visualize the accumulation of balls. 8. **Visualization:** Use a high-contrast color scheme to clearly distinguish between elements. Specifically, use yellow for the structural elements (walls, top guides, physical bin dividers, ground), a contrasting color (like red) for the pegs, and a highly contrasting color (like dark grey or black) for the balls. 9. **Demonstration:** The simulation should visually demonstrate the formation of the normal (or binomial) distribution as multiple balls fall through the pegs and collect in the bins. Ensure the physics parameters (restitution, friction, density) and ball drop rate are tuned for a smooth and clear demonstration of the distribution. #OpenAI Sam Altman Greg Brockman AshutoshShrivastava Aidan McLaughlin

Gemini 2.5 Flash demolishes my Galton Board test, I could not get 4omini, 4o mini high, or 03 to produce this. I found that Gemini 2.5 Flash understands my intents almost instantly, code produced is tight and neat. The prompt is a merging of various steps. It took me 5 steps to achieve this in Gemini 2.5 Flash, I gave up on OpenAI models after about half an hour. My iterations are obviously not exact. But people can test with this one prompt for more objective comparison. Please try this prompt on your end to confirm: -------------------------------------------------- Create a self-contained HTML file for a Galton board simulation using client-side JavaScript and a 2D physics engine (like Matter.js, included via CDN). The simulation should be rendered on an HTML5 canvas and meet the following criteria: 1. Single File: All necessary HTML, CSS, and JavaScript code must be within this single `.html` file. 2. Canvas Size: The overall simulation area (canvas) should be reasonably sized to fit on a standard screen without requiring extensive scrolling or zooming (e.g., around 500x700 pixels). 3. Physics: Utilize a 2D rigid body physics engine for realistic ball-peg and ball-wall interactions. 4. Obstacles (Pegs): Create static, circular pegs arranged in full-width horizontal rows extending across the usable width of the board (not just a triangle). The pegs should be small enough and spaced appropriately for balls to navigate and bounce between them. 5. Containment: * Include static, sufficiently thick side walls and a ground at the bottom to contain the balls within the board. * Implement physical static dividers between the collection bins at the bottom. These dividers must be thick enough to prevent balls from passing through them, ensuring accurate accumulation in each bin. 6. Ball Dropping: Balls should be dropped from a controlled, narrow area near the horizontal center at the top of the board to ensure they enter the peg field consistently. 7. Bins: The collection area at the bottom should be divided into distinct bins by the physical dividers. The height of the bins should be sufficient to clearly visualize the accumulation of balls. 8. Visualization: Use a high-contrast color scheme to clearly distinguish between elements. Specifically, use yellow for the structural elements (walls, top guides, physical bin dividers, ground), a contrasting color (like red) for the pegs, and a highly contrasting color (like dark grey or black) for the balls. 9. Demonstration: The simulation should visually demonstrate the formation of the normal (or binomial) distribution as multiple balls fall through the pegs and collect in the bins. Ensure the physics parameters (restitution, friction, density) and ball drop rate are tuned for a smooth and clear demonstration of the distribution. #OpenAI Sam Altman Greg Brockman AshutoshShrivastava Aidan McLaughlin

RameshR

247,923 Aufrufe • vor 1 Jahr

I'm oddly proud and stoked about this one - live video preview w/ audio of the Robotron AI training on the web. You can watch it live here: I started simple yesterday, a canvas I updated with a snapshot of the game that I pumped over a socket from an emulator session. That evolved into a video tag with a properly encoded video stream, and today I spent a full damned day almost adding audio and keeping it in sync with the game (which runs at an uncorked and unpredictable framerate). Assuming my machine can handle the load, it's available for viewing pleasure now at the link above :-) You have to click on the preview to enable the audio, and you can turn it off/on at the bottom of the page. Thanks to dsobeski for beta-testing it!

I'm oddly proud and stoked about this one - live video preview w/ audio of the Robotron AI training on the web. You can watch it live here: I started simple yesterday, a canvas I updated with a snapshot of the game that I pumped over a socket from an emulator session. That evolved into a video tag with a properly encoded video stream, and today I spent a full damned day almost adding audio and keeping it in sync with the game (which runs at an uncorked and unpredictable framerate). Assuming my machine can handle the load, it's available for viewing pleasure now at the link above :-) You have to click on the preview to enable the audio, and you can turn it off/on at the bottom of the page. Thanks to dsobeski for beta-testing it!

Dave W Plummer

15,227 Aufrufe • vor 4 Monaten

here's what i vibecoded today: punchingface 🥊 an app to make Hugging Face models fight each other on coding and canvas challenges, built with qwen3.6 35b a3b in 24 hours! benchmarks numbers don't mean anything anymore, we need a way to visualize what the models are actually capable of, and canvas are one the best way to showcase it imo. why? because one single error in the code and everything breaks it shows the differences in a matter of seconds, way easier than manually reviewing the quality of the code produced for a complex project; much needed in the space with all the new finetunes dropping everyday! i did a little demo here with qwen3.6 vs qwopus glm 18b merged, the frankenstein model from Kyle Hessling the winner is clear here, qwen is crazy good and has nothing to prove. that said, qwopus 18b isn't terrible at all; the result isn’t the prettiest to the eye, but hey… it works! i've seen so many models just output a blank page (completely non-working code) so this is already a win frankenstein talks and thinks but he needs some extra brain surgery 🧠 results were expected (it's a very experimental model) but love the effort in the 18b direction from jackrong and kyle! the app was entirely vibecoded with qwen3.6, i didn't edit a single file manually. i can say with confidence that it really has the intelligence of claude sonnet 4.5 at a speed of 125tok/s on an rtx 5080 which models should i make fight next?

here's what i vibecoded today: punchingface 🥊 an app to make Hugging Face models fight each other on coding and canvas challenges, built with qwen3.6 35b a3b in 24 hours! benchmarks numbers don't mean anything anymore, we need a way to visualize what the models are actually capable of, and canvas are one the best way to showcase it imo. why? because one single error in the code and everything breaks it shows the differences in a matter of seconds, way easier than manually reviewing the quality of the code produced for a complex project; much needed in the space with all the new finetunes dropping everyday! i did a little demo here with qwen3.6 vs qwopus glm 18b merged, the frankenstein model from Kyle Hessling the winner is clear here, qwen is crazy good and has nothing to prove. that said, qwopus 18b isn't terrible at all; the result isn’t the prettiest to the eye, but hey… it works! i've seen so many models just output a blank page (completely non-working code) so this is already a win frankenstein talks and thinks but he needs some extra brain surgery 🧠 results were expected (it's a very experimental model) but love the effort in the 18b direction from jackrong and kyle! the app was entirely vibecoded with qwen3.6, i didn't edit a single file manually. i can say with confidence that it really has the intelligence of claude sonnet 4.5 at a speed of 125tok/s on an rtx 5080 which models should i make fight next?

left curve dev

19,114 Aufrufe • vor 2 Monaten

“blank canvas problem” figma says raw and unformed feelings… this should not be shared. * Look, I love AI! I lead and design products for AI & Data Solutions in one of the largest firms in the world. I use these tools every day. I’m one of the biggest advocates of AI. I do NOT fear it. I leverage it. I appreciate what Figma is trying to do here. Good intentions to help designers build faster. Make the design easier and accessible for everyone. And of course, meanwhile, profit from It. it is a business after all and it is perfectly fine. While I was watching Dylan generate detailed designs from a prompt, I started seeing aspiring designers getting attached to these tools and having everything easy, ready, quick… and a series of questions started to arise in my mind; * ➟ Are we going to see a generation of designers who lack resilience and love/appreciation for craft? ➟ Designers who are easily hurt and seek help from AI tools every time a client raises a criticism for the proposed design? ➟ Are these designers going to blame AI tools for the outcome and not take accountability for it? ➟ Perhaps they will go back to the AI input and generate 20+ options for clients in 5 mins without thinking instead of talking to the client, understanding the product positioning and actually thinking/exploring/working for the best design solution? ➟ Are we going to see lazier designers? and weak, giving up, getting frustrated easily? * Dylan says ”This tool helps us to get past the blank canvas problem” Blank canvas is one of the best parts of this work. A new fresh start with endless possibilities. Staring at the blank screen and reflecting on our past experiences, our conversations with the client, transforming these thoughts on screen pixel by pixel… That’s how we truly understand the problem at hand and grow as designers. You may say I’m being a romantic, trying to stay in the past. No. Again, I build these tools and encourage everyone to leverage them. However, yes, I have concerns. Not about the efficiencies this technology brings, Concerns about; AI may raise a generation of weak-minded designers who lack an appreciation for the craft. * It is 2:14 am, I should go to sleep. I should not share this writing. this is not meant to be for others but for me. just thinking in writing… thinking…. thinking… …

“blank canvas problem” figma says raw and unformed feelings… this should not be shared. * Look, I love AI! I lead and design products for AI & Data Solutions in one of the largest firms in the world. I use these tools every day. I’m one of the biggest advocates of AI. I do NOT fear it. I leverage it. I appreciate what Figma is trying to do here. Good intentions to help designers build faster. Make the design easier and accessible for everyone. And of course, meanwhile, profit from It. it is a business after all and it is perfectly fine. While I was watching Dylan generate detailed designs from a prompt, I started seeing aspiring designers getting attached to these tools and having everything easy, ready, quick… and a series of questions started to arise in my mind; * ➟ Are we going to see a generation of designers who lack resilience and love/appreciation for craft? ➟ Designers who are easily hurt and seek help from AI tools every time a client raises a criticism for the proposed design? ➟ Are these designers going to blame AI tools for the outcome and not take accountability for it? ➟ Perhaps they will go back to the AI input and generate 20+ options for clients in 5 mins without thinking instead of talking to the client, understanding the product positioning and actually thinking/exploring/working for the best design solution? ➟ Are we going to see lazier designers? and weak, giving up, getting frustrated easily? * Dylan says ”This tool helps us to get past the blank canvas problem” Blank canvas is one of the best parts of this work. A new fresh start with endless possibilities. Staring at the blank screen and reflecting on our past experiences, our conversations with the client, transforming these thoughts on screen pixel by pixel… That’s how we truly understand the problem at hand and grow as designers. You may say I’m being a romantic, trying to stay in the past. No. Again, I build these tools and encourage everyone to leverage them. However, yes, I have concerns. Not about the efficiencies this technology brings, Concerns about; AI may raise a generation of weak-minded designers who lack an appreciation for the craft. * It is 2:14 am, I should go to sleep. I should not share this writing. this is not meant to be for others but for me. just thinking in writing… thinking…. thinking… …

Oykun

38,358 Aufrufe • vor 2 Jahren

I built a multiplayer embroidery sampler for the internet. It's called Common Thread. 🪡 Built the entire thing in Figma Make, first thing I've ever vibecoded with it. Every visitor gets a patch on a shared canvas. You pick a thread color, choose a stitch type, and craft your art. A sampler is a piece of fabric where you practice and collect different stitches — a reference guide and playground in one. They were meant to be passed between hands and added to over time. Common Thread is that, but the fabric is infinite, the UI is whimsical, and the hands are anonymous. The part I love most: you can snap any section with a draggable embroidery hoop and export it, framed. i love how Figma Make is a kind of tool that lets you focus on what something should feel like, not how to ship it. Made for the Contra x Figma Makethon #FigmaMakeathon Try it yourself! /link below/

I built a multiplayer embroidery sampler for the internet. It's called Common Thread. 🪡 Built the entire thing in Figma Make, first thing I've ever vibecoded with it. Every visitor gets a patch on a shared canvas. You pick a thread color, choose a stitch type, and craft your art. A sampler is a piece of fabric where you practice and collect different stitches — a reference guide and playground in one. They were meant to be passed between hands and added to over time. Common Thread is that, but the fabric is infinite, the UI is whimsical, and the hands are anonymous. The part I love most: you can snap any section with a draggable embroidery hoop and export it, framed. i love how Figma Make is a kind of tool that lets you focus on what something should feel like, not how to ship it. Made for the Contra x Figma Makethon #FigmaMakeathon Try it yourself! /link below/

charlota

519,718 Aufrufe • vor 4 Monaten

🚨 How players run when the connection is good EA needs to fix the server issues. Regarding AI defending, count the number of blocks and interceptions made by non-controlled players during FC PRO, then compare it to online matches with a bad connection. Also, look at the end of the video and see what Theo is doing. He moves away from the striker even though there is no threat on the wing. In a bad connection situation, if your attack is delayed by just one second, what should have been a penalty won't be given, and it would result in an interception instead. Don't get me wrong, I don't think the game is perfect. I hate tackling from behind, the L1 boost skyrocketing, and first-time through balls, but I can't defend a lie. The "Shadow Patch" theory is a lie. The servers are terrible. The excessive number of animations in attack is also terrible. We want a simple, fast, and fun game. We don't need millions of animations that interact poorly on these overloaded servers.

🚨 How players run when the connection is good EA needs to fix the server issues. Regarding AI defending, count the number of blocks and interceptions made by non-controlled players during FC PRO, then compare it to online matches with a bad connection. Also, look at the end of the video and see what Theo is doing. He moves away from the striker even though there is no threat on the wing. In a bad connection situation, if your attack is delayed by just one second, what should have been a penalty won't be given, and it would result in an interception instead. Don't get me wrong, I don't think the game is perfect. I hate tackling from behind, the L1 boost skyrocketing, and first-time through balls, but I can't defend a lie. The "Shadow Patch" theory is a lie. The servers are terrible. The excessive number of animations in attack is also terrible. We want a simple, fast, and fun game. We don't need millions of animations that interact poorly on these overloaded servers.

FUT POLICE LEAKS

86,653 Aufrufe • vor 1 Jahr