Only if education could be this interactive ❤️‍🔥 I've had a looong wish to build something genuinely useful through vibe coding, and I finally did it. A 3D human anatomy application built with @threejs using GPT 5.6 Sol. It all started with a single design image that I created using GPT Image 2.0. I then used it to generate every 3D organ image, one by one. Next, I converted each of those images into 3D models using @tripoai (and no, they didn't sponsor this 😄). After that, I opened Codex, wrote a master prompt based on the design, and gave it the prompt, the design image, and all the 3D models. Codex built the first version beautifully, but there was one big problem. Every single 3D model was nearly 120-150 MB. That obviously wasn't practical for the web and was giving a performance of 16fps. After a few iterations, Codex optimized each model down to roughly 2–5.5 MB while preserving the visual quality, reducing the total asset size from ~900 MB to just 28.6 MB. And each model loads on demand. Along the way, Codex also generated those anatomical illustrations showing where each organ sits in the human body, and even created the interactive hotspot markers that explain different parts of every organ. It handled all of that. The process wasn't exactly one shot, but it also wasn't difficult. You just have to do it step by step. It genuinely felt like building something that could make learning anatomy much more engaging. The inspiration came from @DilumSanjaya's 3D animal plant cell project. I remember seeing it and thinking, "I want to build something like this one day." And I did it :D Live: anatomy-livid.vercel.app/ Code: github.com/thebuggeddev/anat…

Honestly, I wasn't expecting Gemini 4 Argon to go this far with a circus tent. All I wanted was to test how well it could handle fabric physics. And it ended up creating such a realistic 3D procedural bmesh geometry with custom real time cloth & artillery physics for whole circus. Really impressed with this one 🎪🔥

Nope, that’s not Apple’s website 😄 It’s built with Gemini 4 Argon, featuring custom 3D HomePod models so real to original,.. even Apple doesn’t have them :D

If Earth were tilted 90° instead of 23.4°, you'd be sipping cold drinks during a warm Arctic summer and melting in London at 41°C in July 🥵 Kinda crazy to think about, but yes. I've always wondered how we're living in such a tiny window of possibility on this huge ball floating in the vast emptiness of space. So, I built a 3D Earth simulator where you can tilt the planet from 0° to 90° and watch how the seasons change for every month and even every date. You can tap any country to see its temperature and daylight hours throughout the year. It's all based on real climate data and real physics. Built with Claude Opus 5.5.

Couldn’t ask for anything better than Derya appreciating my little work 🥹 And look at that, you can build so many amazing things using Gemini 4 Argon.

Working with 3D worlds is getting stupid easy these days, huh? :D

I gave Gemini 4 Argon one hell of a challenge. To Rebuild the Cathedral of Football, Wembley Stadium, in the browser 🏟️ And result is a fully interactive 3D stadium with 90,360 individually clickable seats, allowing you to select any Block, Row and Seat and fly straight to that exact viewing position to experience the pitch from your seat before booking a ticket. Built with Three.js WebGPU and TSL, it runs at 60 FPS with custom shaders, an intricate stadium roof and Wembley’s iconic arch. Sure, it's not perfectly exact but should be near possible. I got here in just a couple of prompts so with more refinement, I genuinely think this could get incredibly close to the real thing. Also, I’ve experimented with football builds using Claude Fable 5 and GPT 6 Astra before too, and honestly, this result puts Gemini 4 Argon firmly on my frontier list. It's doing so well!

What if puzzles could be both fun and educational? 🧠 I came across these human body brain puzzles on Pinterest and loved it so much that I immediately decided to give it a try. And here’s the result, created entirely with Gemini 4 Argon!! Wouldn't it be amazing to have such website that teaches different concepts through fun, interactive puzzles? Might make learning feel less like studying and more like playing :D

Not sure what you’re not seeing in Argon, but it’s so good I can’t stop using it. Asked it to create a Warlord’s Throne, and it generated this entire customizable 3D model. The cloth textures, bull horns, and intricate details are SO well done 🔥

From using Gemini as my first AI model to learn, experiment, and build with, to having Google feature my work. 🥹❤️ This genuinely means so much to me. Thank you so so much, Google!!

You can do SO much more with Gemini 4 Argon 🫠 Asked it to build an interactive website to teach electric field lines in action, and it did so in ONE prompt, complete with challenges and dedicated learning pages. I barely even explained how to build whole functionality. Argon just figured it all out. The level of intelligence we have at our fingertips these days,.. damn how so amazing!!

You’re seriously underestimating what Gemini 4 Argon is capable of It didn’t just create another procedural fluffy 3D spider plushie GLB but also rigged the entire thing with realistic 8-leg physics, ground-aware foot placement, and natural body movement. You can literally pet it, watch it react, and make it hop around. All without touching Blender. 🕷️ This time, it’s even more detailed, with 65,000 fur strands and intricate embroidered features, all built using Three.js and GLSL shaders. I’ve been experimenting with 3D model creation using Argon, and the results just keep getting crazier with every build. You really need to try Argon!!

A lot of people, including Theo, have called Gemini 4 Argon unusable for real code work. So I asked it to refactor the entire Gothic Victorian Carriage 3D website it built and make whatever improvements it thought were necessary. Entirely on its own. 🫡 It literally turned a 4,290 line monolithic HTML file into modular ES6 architecture, replaced 25MB of blocking base64 JavaScript with binary glb streaming, eliminated per-frame Vector3 allocations and DOM layout reads, and shared GPU buffers across wheel pairs. All without changing a single pixel, that too in one attempt. You can see performance below. I'm not sure how everyone is using Argon but it really isn't "unusable" lol Btw, you can check whole codebase here, open sourced: github.com/thebuggeddev/carr…

Someone asked if Gemini 4 Argon is actually good enough to customize every aspect of a 3D model, not just generate one. Well, here's Argon doing the talking. You can tweak carriage's dimensions live, with the chassis, axles & leaf springs adapting automatically. Customize wheels, rims, spokes, working doors, Gothic windows, roof shape, suspension, trims, lanterns & paint. EVERYTHING is customizable. 🤯

I feel like we haven't seen enough of what Gemini 4 Argon can actually do ❤️‍🔥 I gave it one image of a Gothic Victorian carriage and asked it to build the whole thing as an interactive 3D experience. Just one image and this is what you get. A 174,000 face model from scratch, complete with opening doors, hinged windows, working wheel physics, suspension, flickering lanterns and even custom sculpted skulls. All of this generated through code. No blender. Argon absolutely delivered with this one too. Many more such experiments are on the way :D

Gemini 4 Argon, everyone!!! 🤯 It literally created a ridiculously realistic 3D fluffy pumpkin plushie without using Blender at all. In fact, it generated it as a procedural GLB model with 28 volumetric fur shells, 60,000 GPU-instanced 3D fur strands, and even real-time air-jet physics, all built with Three.js and GLSL. You can literally blow the fur around in real time. 💨 Totally nuts.

Gave Claude a photo of packaging concept and asked it to recreate it in 3D. It built that entire carton using math, including curved panels, corrugated edges, transparent windows, and a folding lid. Made with Three.js and WebGPU by Claude Opus 5.5

If you want to learn how to build 3D websites with AI without losing your mind in the process, this workshop is for you. discord.gg/Jpj5GXghq?event=1… P.S. You’ll also get FREE Tripo credits just for showing up 👀

You have to try this skill ❤️‍🔥 I tried recreating a watermill machine with it, and I’m genuinely impressed by how neatly it got everything right in one shot. Amazing!! It did take 750K+ tokens though, that too on Claude Opus 5.5 Max :)

Mistral feels like ChatGPT in its early days 🥴 Tested it on a pretty basic dashboard, and it completely went in its own direction, damn. It actually created a very generic template instead of following the design it was given and was strictly asked to follow.

You’ve seen me build all kinds of 3D websites, and many of you have asked how to create something like that yourself 👀 So we’re making it happen I’m teaming up with @tripoai and @NickDevFE, the creator of img2threejs, for a live workshop where we’ll show you exactly how to do it. Here’s what you'll learn 👇 ✴️ Generate detailed, high-quality 3D characters with Tripo AI ✴️ Reconstruct 3D models as editable Three.js code using img2threejs ✴️ Modify geometry, customize individual components, and refine model details ✴️ Rig characters, create, repair AI generated animations, and VFX ✴️ Integrate everything into an immersive, interactive 3D website using Three.js And if you attend, there’s something FREE waiting for you, along with another surprise we think you’ll love 💛 🗓️ October 12 at 14:00 UTC 📍 Live on Tripo AI’s Discord Join us: discord.gg/84xJc7QzW And if there’s anything you’d like me to cover during the workshop, let me know below :)