Fredrik Ehne

昔のパソコンみたいでめっちゃ可愛くない？

In that split second Enzo Fernandez has only one choice. Yes, he can lose his nation the World Cup. But then he would miss the chance to commit an amazing, amazing, incredible foul. And if you understand Enzo Fernandez, you would understand he never even had a choice to make

Google has solved a 20-year-old web problem.. It's called html-in-canvas, a new experimental Chrome feature that renders live HTML directly as a 3D texture. here's what it does: → draws any DOM element (dashboards, forms, entire app UIs) straight into a canvas → full CSS still works, flexbox, grid, shadows, backdrop-filter, animations → text stays selectable, inputs stay typeable, screen readers still see it → auto syncs with your DOM on every paint → pipes into any WebGL shader → distort, curl, refract, warp the wildest part: someone already built an html-cloth plugin that turns your live UI into simulated fabric.

MCP CAN DRIVE BLENDER ALL DAY AND STILL NOT SCULPT ONE SPINE ON THIS THING a model through MCP writes bpy, blocks out a base mesh, sets the scene, spins the camera, hits render, reads its own frame. that whole loop got cheap. the scaffold of a creature is a conversation now. but look at what's actually on screen. the layered spines. the skin that folds like it has muscle under it. the asymmetry that reads as alive instead of generated. that's sculpt work. hours in the brush, by someone who knows how a jaw should sit and where a ridge breaks. no prompt makes organic detail feel earned. MCP gives you the blockout, the lighting, the turntable. it doesn't give you the hand that pushed every spike. the technical middle of 3d collapsed. the sculpt, the taste, the thing that makes a monster look like it could bite, still didn't move.

HTML isn't just markup anymore. For AI agents, it's becoming the best visualization layer. Made with @HyperFrames_ Shoutout to @equinusocio for the css mask inspo 🔊 Don't mute this

this is the plan for the new data center that Sam Altman is planning to build in Big Sur. How is this allowed?!?!?

Day 7 of building a snowboarding game with Fable 5🏂 It's got a soundtrack now, PS1-style jungle/DnB, and it plays on mobile. Enhanced snow, lighting, and sky quality. New terrain. Vertical stick control. 1 week 42,000 lines of code, All raw Webgpu, no engine.

Imagine the Apple car was made by BRAUN and not by Ferrari (Luce). The BRAUN Auto01 city car. Actually useful.

Snowboarder one-shot on Kimi K3 (OpenCode, Godot). Almost no visual bugs, task cleanly implemented. Better than Sol 5.6 and Opus here. @Kimi_Moonshot

After adding meds, I now added the brain with 222 interactive parts. All mapped to the relevant other 2,277 body parts, 8,382 conditions and 2,776 medications. This is starting to be useful in pattern recognition that goes beyond typical isolated diagnosis. Interesting thing was that I did not decide to add a more detailed brain, Founden AI did overnight.

gm gm ☕️

One of my favorite workflows lately: I create my characters in Midjourney, then make them look real with GPT

Kimi K3 built a fully procedural browser-based 3D exploration game using Three.js WebGPU and GPU compute. It procedurally generated the environment, while using a 3D asset generation tool to create the rider and horse models, producing an expansive open world with forests, a log-cabin village, snowy mountains, and dynamic weather. External assets used: animated cowboy and horse models and terrain data.

很多人对threejs的想法还停留在单纯的3d游戏场景还原. 其实最有创造力的实现，是coding还原真实物理规律 。 我们用Fable+threejs用于极地探险的专业房车设计，但是过去不存在，只存在想象中，我们做了一个物理还原的3d互动游戏，可以拆解不断的零部件，细节，规律，自由组合，游戏就是最好的世界模拟环境。 有其他类似的作品可以分享的吗 #fable @claudeai @threejs #threejs

GOOGLE DEEPMIND JUST TURNED A RANDOM OFFICE WALKTHROUGH INTO A SEARCHABLE 4D WORLD YOU CAN QUERY IN PLAIN ENGLISH. You feed it raw video. Nothing tagged, nothing labeled, no manual annotation. GenCeption tears that footage apart into four layers at once: a depth map, surface normals, a camera raymap, and segmentation masks. Then it rebuilds the whole space as a 4D scene you can actually talk to. The demo is stupid simple and that's the point. Type "I want to find a comfortable place to sit" and the sofa lights up. "I need a whiteboard." Whiteboard, found. "Is there a monitor I can use?" Highlighted. "Where's my colleague?" It grounds the person. No bounding boxes drawn by hand. No labels typed in advance. The model figured out what a sofa IS and where it sits in 3D space, from pixels alone. Every video you've ever recorded is now a database you can search with a sentence. Robots, AR, security, indexing. Same primitive. The camera roll just became queryable.

China open-sourced a model that reconstructs any scene in 3D from a regular video, in real-time. one camera. no LiDAR. 10,000+ frames without falling apart. just walk around with your camera and watch the entire world get rebuilt in 3D at 20 fps. → runs at ~20 FPS on a single GPU → Stable over 10,000+ frames → Beats optimization-based methods on benchmarks → Works on drone footage, driving videos, indoor walkthroughs 100% open source.

You can now vibe code a language model. From a single prompt, GPT‑5.6 built the entire training pipeline and trained a model from scratch on my iMessage history. Locally on my Mac. It now generates replies in my writing style.

I had Fable build another thing I always wanted, a full procedural fantasy kingdom generator with economics, trade routes, population growth, wars, lineages, and occasional dragons. First, I worked with it on a plan, then it made it. You can play it here: annals-kingdom.netlify.app

Been super into 1970s graphic design lately