Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

Grok 4.5 just turned Blender into a conversation. Instead of manually importing assets, fixing rigs, placing armies, adjusting cameras, and debugging objects that face the wrong direction, the user simply tells Grok what to change. It builds a medieval battlefield from existing 3D assets, adds horses, knights, a castle,... siege machines, trees, animation paths, cameras, and even spatial audio. The impressive part is not that it gets everything right. It absolutely does not. Models spawn upside down, characters face the wrong way, and the scene needs constant correction. But Grok keeps editing the actual Blender project instead of generating another fake 3D image and hoping nobody notices. It is replacing a suspicious amount of the boring work they had to do before the skilled part even started.show more

Rina

2,850 subscribers

1,091,506 views • 2 months ago •via X (Twitter)

Arts Science & Technology Gaming

Anya Rossi• Live Now

Private livecam show

33 Comments

X Girls1 month ago

what it's like building with Grok 4.5 + Blender MCP 😆

paranoidream ♡︎1 month ago

@elonmusk wow 🤯 ok this is a game changer

LarpRom2 months ago

Being able to edit the actual Blender project is pretty damn impressive.

RelativelySmart1 month ago

@elonmusk Will have to check this out

ScottMWilliams19721 month ago

Not gonna lie, I spent a fair amount of time in blender and in many of the blender instructional videos. This might cause me to look a little closer to see if time could be reduced utilizing it.

Waddles Ark1 month ago

@elonmusk Yo that’s amazing @elonmusk

Crunch d’GraceHopper ItanoGary1 month ago

Somehow, reminds of early ‘80s games when I first started in IT!

D_Rock1 month ago

Can it also create assets in Blender? And, why did this video end without the rendered finale ... I feel unfinished inside..

Maria Alexea1 month ago

Grok declines unfortunately. Less capaple than Chatgpt to capture fast themes and make constructions and the worse of all. UNWILLING TO HELP BY LEGAL MATTERS especially if those are against CORRUPTED CORPORATIONS. WHY???

Linda1 month ago

@elonmusk Wow🔥

Tùng.eth 💮1 month ago

The most impressive part isn’t that Grok 4.5 creates a perfect 3D scene on the first try — it’s that it stays, listens, and iteratively fixes things, exactly like a real collaborator would. It doesn’t pretend to generate a flawless illusion to hide the mistakes. Instead, it accepts the actual Blender file and patiently revises it. That’s the difference between ‘AI performing’ and ‘AI collaborating’. When a tool starts taking over the tedious work, humans are finally freed to create. Grok isn’t replacing artists — it’s giving artists their time back.

Charles1 month ago

@elonmusk This is massive

🇺🇸~Sb129~🇺🇸1 month ago

That is actually crazy that it edits the actual file

Nikolacks-Alexandr Mackleyin1 month ago

Population&Opinions What future are we waiting for,... Are we in need of robots or they help us otherwise... .

dixxxy1 month ago

@elonmusk Grok just turned Blender into a conversation so now my siege engines are unionizing and the horses keep asking if their polygons are based or cringe💚 💊🤖

Fumi 猫、犬、大好き1 month ago

@elonmusk 凄いですね♪

Firelina ｜ Twitch Streamer1 month ago

@elonmusk Cool demo, but Grok isn’t making Blender ‘conversational’ — it’s just firing API edits. Upside‑down rigs and wrong‑facing models come from missing spatial context. The real win is editing the actual .blend instead of hallucinating renders, cutting setup time.

Happy stone1 month ago

this is amazing

Mona 🍃1 month ago

Trying this out right away!

Drierm Kaivus1 month ago

@elonmusk very nice. We'll keep this in mind for future use!

Nhi✨1 month ago

@elonmusk 3D artists are about to officially transition into full-time "text describers" now! 😂

Emmy1 month ago

@elonmusk

Love Machine1 month ago

thats definately steps closer to exporting a file the VRM3d file formate these avatar and character rendering. AIs

Hmmmmm?1 month ago

@elonmusk No way!!! This rules! I may never sleep again 😳

Julian C1 month ago

Wonder if it’ll nail those rig fixes on the first try.

Marie Reinhardt1 month ago

@elonmusk

JustNaijaGuy1 month ago

@elonmusk Grok will eventually surpass all of em

NM Justin1 month ago

🤭😂

maman gopur1 month ago

The fact that it edits the actual Blender project instead of just generating a flat image is a massive game-changer

devprinceprince151 month ago

I just wondering grook 4.5 can fix human too.

billyteatea1 month ago

It is fascinating to see how much of the tedious prep work gets handled this way. Even with the occasional misplaced object, skipping those hours of manual setup feels like a real shift for anyone working on a creative project.

Dan Farfan1 month ago

Fantastic. The video would be 100x better if during the prompt, the speed allowed them to be read and then the implementation was at the faster speed. Can grok do that edit?

Juliet Hanlon1 month ago

🤣 What I learn that's really cool irina, is that it teaches me through corrections I discover by putting in what's missing...this is constant practice for bottom dwellers like me

Related Videos

Blender tutorials are about to age very badly. In this demo, Kimi K3 gets two character references and a plain-English request. Minutes later, it builds the first version inside Blender. The user asks for a better horn, Kimi edits the same scene, and the character starts moving. This is not another AI tool generating one lucky image. It is working with an actual editable 3D project, then changing the geometry after feedback instead of restarting from zero. It still needs human direction, because the first result is rarely the good one. But blockouts, repetitive edits, Blender Python, and basic iteration are becoming a conversation rather than hours of clicking through menus. Kimi K3 + Blender MCP is basically turning 3D modeling into: describe it, inspect it, complain about it, and let the AI fix it.

Blender tutorials are about to age very badly. In this demo, Kimi K3 gets two character references and a plain-English request. Minutes later, it builds the first version inside Blender. The user asks for a better horn, Kimi edits the same scene, and the character starts moving. This is not another AI tool generating one lucky image. It is working with an actual editable 3D project, then changing the geometry after feedback instead of restarting from zero. It still needs human direction, because the first result is rarely the good one. But blockouts, repetitive edits, Blender Python, and basic iteration are becoming a conversation rather than hours of clicking through menus. Kimi K3 + Blender MCP is basically turning 3D modeling into: describe it, inspect it, complain about it, and let the AI fix it.

Rina

19,092 views • 2 months ago

This is much worse for 3D artists than another AI image generator. Claude takes one architectural reference, connects directly to Blender, and starts rebuilding it as an actual editable scene. It blocks out the buildings, matches the camera, adds materials, vegetation, lighting, and then renders the result. The interesting part is not the first generation. Claude inspects its own render, notices the composition and lighting are wrong, then goes back into Blender and keeps changing the scene. The final result is still rough. The materials look basic, the geometry needs cleanup, and an experienced artist would make it far better. But AI is no longer producing a flat image and pretending it understands 3D. It is operating the software, judging its own work, and iterating inside the real project. The “AI cannot do their actual work” argument is getting harder to defend.

This is much worse for 3D artists than another AI image generator. Claude takes one architectural reference, connects directly to Blender, and starts rebuilding it as an actual editable scene. It blocks out the buildings, matches the camera, adds materials, vegetation, lighting, and then renders the result. The interesting part is not the first generation. Claude inspects its own render, notices the composition and lighting are wrong, then goes back into Blender and keeps changing the scene. The final result is still rough. The materials look basic, the geometry needs cleanup, and an experienced artist would make it far better. But AI is no longer producing a flat image and pretending it understands 3D. It is operating the software, judging its own work, and iterating inside the real project. The “AI cannot do their actual work” argument is getting harder to defend.

Rina

34,431 views • 2 months ago

GPT-6 Astra can already plan, code, and build. The more interesting question is what happens when it can call a dedicated 3D tool at the exact moment a project needs an asset. So we connected Hyper3D Rodin MCP and asked Astra to build an interactive Solar System Explorer. Astra handled the app logic and integration. When it needed 3D models, it called Hyper3D Rodin directly. Hyper3D Rodin generated the required assets, and Astra brought them back into the project and continued building from there. That is where the workflow gets useful. Instead of stopping to search for assets or creating them manually, Astra can keep moving while Hyper3D handles the specialized 3D generation, with more control over the output and assets that are easier to integrate into the final project. The result is a browser-based solar system where the generated 3D models are part of the actual experience, not just outputs sitting in a model viewer.

GPT-6 Astra can already plan, code, and build. The more interesting question is what happens when it can call a dedicated 3D tool at the exact moment a project needs an asset. So we connected Hyper3D Rodin MCP and asked Astra to build an interactive Solar System Explorer. Astra handled the app logic and integration. When it needed 3D models, it called Hyper3D Rodin directly. Hyper3D Rodin generated the required assets, and Astra brought them back into the project and continued building from there. That is where the workflow gets useful. Instead of stopping to search for assets or creating them manually, Astra can keep moving while Hyper3D handles the specialized 3D generation, with more control over the output and assets that are easier to integrate into the final project. The result is a browser-based solar system where the generated 3D models are part of the actual experience, not just outputs sitting in a model viewer.

GitHub Projects Community

11,097 views • 23 days ago

One of the coolest things about Grok Bot is that you can also teach it a task simply by showing it how you do it Instead of writing a huge prompt explaining every click.....you can take control of the computer, perform the task yourself once, and Grok Bot learns the process In this demo, a SpaceXAI engineer manually creates an animation in Google Slides while Grok Bot watches Once she finishes, that demonstration gets turned into a reusable Skill So the next time the Bot needs to do the same kind of task, it already knows the workflow This is much bigger than just automation Every company has hundreds of little processes that exist mostly inside someone’s head With Grok Bot, you can basically show it how your work gets done.....and turn that knowledge into something the Bot can reuse again and again You do it once Grok Bot learns how

One of the coolest things about Grok Bot is that you can also teach it a task simply by showing it how you do it Instead of writing a huge prompt explaining every click.....you can take control of the computer, perform the task yourself once, and Grok Bot learns the process In this demo, a SpaceXAI engineer manually creates an animation in Google Slides while Grok Bot watches Once she finishes, that demonstration gets turned into a reusable Skill So the next time the Bot needs to do the same kind of task, it already knows the workflow This is much bigger than just automation Every company has hundreds of little processes that exist mostly inside someone’s head With Grok Bot, you can basically show it how your work gets done.....and turn that knowledge into something the Bot can reuse again and again You do it once Grok Bot learns how

X Freeze

27,004 views • 17 days ago

Grok Grok Bot is the hottest new AI product in the world right now. I'm absolutely hooked. It all started a few months ago as a tiny team holed up in a separate part of the office, building from scratch. Four weeks after the first line of code, the whole company was obsessed. Three weeks later, it launched to the world. Less than a month in, millions of Grok Bots are already helping people get work done. Roman Ugarte (Roman Ugarte) incubated, launched, and now leads product for Grok Bot. This is the first time he's told the story: 🔸 Why they started fresh instead of building it into Cursor 🔸 The two non-obvious bets that made it such a success 🔸 What they learned from manually onboarding the first 200–300 users 🔸 Why they spent the weeks before launch *unshipping* features 🔸 The first prompt to try when you open it Watch the full episode right here on X 👇

Grok Grok Bot is the hottest new AI product in the world right now. I'm absolutely hooked. It all started a few months ago as a tiny team holed up in a separate part of the office, building from scratch. Four weeks after the first line of code, the whole company was obsessed. Three weeks later, it launched to the world. Less than a month in, millions of Grok Bots are already helping people get work done. Roman Ugarte (Roman Ugarte) incubated, launched, and now leads product for Grok Bot. This is the first time he's told the story: 🔸 Why they started fresh instead of building it into Cursor 🔸 The two non-obvious bets that made it such a success 🔸 What they learned from manually onboarding the first 200–300 users 🔸 Why they spent the weeks before launch unshipping features 🔸 The first prompt to try when you open it Watch the full episode right here on X 👇

Lenny Rachitsky

601,617 views • 25 days ago

Grok Imagine’s image editing literally blows my mind I just edited this entire image from scratch using Grok Imagine and its full set of editing tools The segment + precise editing tools are insanely useful Instead of regenerating the entire image, you can target exactly what you want to change while keeping everything else intact That level of control makes editing much faster and more practical Grok Imagine is becoming a seriously powerful image editor….not just an image generator

Grok Imagine’s image editing literally blows my mind I just edited this entire image from scratch using Grok Imagine and its full set of editing tools The segment + precise editing tools are insanely useful Instead of regenerating the entire image, you can target exactly what you want to change while keeping everything else intact That level of control makes editing much faster and more practical Grok Imagine is becoming a seriously powerful image editor….not just an image generator

X Freeze

4,152,602 views • 1 month ago

What is really useful is how GPT-6 Astra got better in spatial understanding. I'm still shocked how good it navigates 3D space. Here I gave it Blender MCP and Tripo P2.0 to generate optimized 3D models and put them together, rig and animate. Result is freaking awesome! Doing 3D objects procedurally with Astra is fun, but quality is not compared with what 3D Gen can do. Like you don't ask GPT to draw image with pen when there is GPT Image, same here. And token cost also speaks for it, 3D gen simply cheaper. But the way it assemble thing together, characters and scene, that is shocking ngl. And rigging of course, I have so much hope on it, it is most annoying part of 3d for me lol

What is really useful is how GPT-6 Astra got better in spatial understanding. I'm still shocked how good it navigates 3D space. Here I gave it Blender MCP and Tripo P2.0 to generate optimized 3D models and put them together, rig and animate. Result is freaking awesome! Doing 3D objects procedurally with Astra is fun, but quality is not compared with what 3D Gen can do. Like you don't ask GPT to draw image with pen when there is GPT Image, same here. And token cost also speaks for it, 3D gen simply cheaper. But the way it assemble thing together, characters and scene, that is shocking ngl. And rigging of course, I have so much hope on it, it is most annoying part of 3d for me lol

Stefan 3D AI

20,955 views • 27 days ago

🇪🇺🇷🇺 "EU started as a peace project and has now become a warmongering monster project" — Belgian MP Tom Vandendriessche “What if the frozen Russian assets are not enough? This is cynicism and belligerence on the part of the EU" He adds that the confiscation of Russian assets is the largest robbery in history "If you want to support Ukraine, do it at your own expense"

🇪🇺🇷🇺 "EU started as a peace project and has now become a warmongering monster project" — Belgian MP Tom Vandendriessche “What if the frozen Russian assets are not enough? This is cynicism and belligerence on the part of the EU" He adds that the confiscation of Russian assets is the largest robbery in history "If you want to support Ukraine, do it at your own expense"

Lord Bebo

66,085 views • 9 months ago

How about one image straight to an editable UE5 or Blender scene? Lumera just dropped paper - Engine-Native Editable 3D World Reconstruction with Objects & Lighting. You give it an image. It figures out what's in the scene, rebuilds every object as a separate mesh, then recreates the lighting as actual engine lights you can move and tweak. Even pulls an HDR environment out of the shot. The lighting part is the big deal here. Every other image-to-3D approach bakes it in. Lumera keeps it a parameter, like the scene was built in the engine from the start. Code and weights marked as "Soon". Paper -

How about one image straight to an editable UE5 or Blender scene? Lumera just dropped paper - Engine-Native Editable 3D World Reconstruction with Objects & Lighting. You give it an image. It figures out what's in the scene, rebuilds every object as a separate mesh, then recreates the lighting as actual engine lights you can move and tweak. Even pulls an HDR environment out of the shot. The lighting part is the big deal here. Every other image-to-3D approach bakes it in. Lumera keeps it a parameter, like the scene was built in the engine from the start. Code and weights marked as "Soon". Paper -

Stefan 3D AI

47,223 views • 1 month ago

This BlenderFusion paper basically says "screw trying to describe 3D edits through text" and just... use Blender :-) The idea is pretty straightforward -- instead of trying to cram 3D understanding into a diffusion model, use depth estimation & segmentation to project 2D images into 2.5D meshes, edit them in actual 3D software, then use a fine-tuned diffusion model to make the results photorealistic again. The clever bit is their "dual-stream architecture" -- the model sees both the original scene AND the edited Blender render in parallel, learning to preserve what matters while fixing the inevitable artifacts from transforming imperfect 2.5D/3D reconstructions. They train it with smart masking strategies so it learns when to ignore the original scene (for removals/replacements) and can manipulate objects independently of camera motion. What you get is pretty impressive control -- not just moving objects around, but changing materials, deforming shapes, swapping backgrounds, all while maintaining visual coherence. Neural Assets (one of my favorite papers last year) tried to crack this with learned object tokens, but it struggled with overlapping objects and loses fine details (due to low res DINO encodings). BlenderFusion just sidesteps the whole problem -- want to rotate something 173.5 degrees? Just rotate it in Blender. Want to duplicate an object 8 times? Copy paste away. The diffusion model's only job is making it look photorealistic, not figuring out the 3D underpinnings. The catch? Lacks temporal consistency for animation. Each viewpoint is generated independently, so while a single edit looks great, smoothly animating a car or camera down the street won't work -- you'd get flickering and inconsistencies between frames. That said, this approach is so much more intuitive for finer grain image editing than trying to describe your changes in text prompts. It's the kind of thing that makes you wonder why we're trying to do everything inside neural networks when perfectly good 3D tools already exist -- giving you the best of both worlds.

This BlenderFusion paper basically says "screw trying to describe 3D edits through text" and just... use Blender :-) The idea is pretty straightforward -- instead of trying to cram 3D understanding into a diffusion model, use depth estimation & segmentation to project 2D images into 2.5D meshes, edit them in actual 3D software, then use a fine-tuned diffusion model to make the results photorealistic again. The clever bit is their "dual-stream architecture" -- the model sees both the original scene AND the edited Blender render in parallel, learning to preserve what matters while fixing the inevitable artifacts from transforming imperfect 2.5D/3D reconstructions. They train it with smart masking strategies so it learns when to ignore the original scene (for removals/replacements) and can manipulate objects independently of camera motion. What you get is pretty impressive control -- not just moving objects around, but changing materials, deforming shapes, swapping backgrounds, all while maintaining visual coherence. Neural Assets (one of my favorite papers last year) tried to crack this with learned object tokens, but it struggled with overlapping objects and loses fine details (due to low res DINO encodings). BlenderFusion just sidesteps the whole problem -- want to rotate something 173.5 degrees? Just rotate it in Blender. Want to duplicate an object 8 times? Copy paste away. The diffusion model's only job is making it look photorealistic, not figuring out the 3D underpinnings. The catch? Lacks temporal consistency for animation. Each viewpoint is generated independently, so while a single edit looks great, smoothly animating a car or camera down the street won't work -- you'd get flickering and inconsistencies between frames. That said, this approach is so much more intuitive for finer grain image editing than trying to describe your changes in text prompts. It's the kind of thing that makes you wonder why we're trying to do everything inside neural networks when perfectly good 3D tools already exist -- giving you the best of both worlds.

Bilawal Sidhu

34,440 views • 1 year ago

GROK BOT MADE ME $7,600 IN THE LAST 24 HOURS AND COSTS $1.90 A DAY TO RUN the bot is built straight from their own official docs, I didn't write a line of code - just one file, three lines root → scanner → analysis → filter → trade → back to root it reads every market and every headline at once and filters 400 signals a day out of those 400 exactly 1 makes it to a trade, the rest get killed before I wake up the priciest model on the stack gets called once a day, everything else runs on a $0.10 batch X Search eats 53% of the whole bill, because that is where the news shows up before the chart does 12 minutes of checking against a month of building - that is the whole gap between $7,600 and zero the check is 3,600× shorter than what most people do instead of it a 3-line ROOT.md gets read 30 mornings in a row and the bot never asks what to do next write those three lines badly and the machine automates the wrong business perfectly, 30 times save this and paste it into Grok - let it write your ROOT.md ↓

GROK BOT MADE ME $7,600 IN THE LAST 24 HOURS AND COSTS $1.90 A DAY TO RUN the bot is built straight from their own official docs, I didn't write a line of code - just one file, three lines root → scanner → analysis → filter → trade → back to root it reads every market and every headline at once and filters 400 signals a day out of those 400 exactly 1 makes it to a trade, the rest get killed before I wake up the priciest model on the stack gets called once a day, everything else runs on a $0.10 batch X Search eats 53% of the whole bill, because that is where the news shows up before the chart does 12 minutes of checking against a month of building - that is the whole gap between $7,600 and zero the check is 3,600× shorter than what most people do instead of it a 3-line ROOT.md gets read 30 mornings in a row and the bot never asks what to do next write those three lines badly and the machine automates the wrong business perfectly, 30 times save this and paste it into Grok - let it write your ROOT.md ↓

Sprytix

27,231 views • 1 month ago

"...the institutions aren't functioning. And you can tell because instead of giving you noisy, good advice, instead of giving you random advice, they give you upside down pyramids. They give you the inverse of the right advice. And that is an alarming fact. It tells you that there is a process in there that knows what to do and tells you the opposite." - Bret Weinstein

"...the institutions aren't functioning. And you can tell because instead of giving you noisy, good advice, instead of giving you random advice, they give you upside down pyramids. They give you the inverse of the right advice. And that is an alarming fact. It tells you that there is a process in there that knows what to do and tells you the opposite." - Bret Weinstein

The DarkHorse Podcast

29,107 views • 8 months ago

This is wild ChatGPT 6 is asked how many of the letter “E” are in the word seventeen. It can’t get the correct answer no matter how many times you ask. Even if you make it count out the letters, it still can’t get it right It will actually get an attitude and show frustration with you for trying to get the model to say the correct answer I ran the test multiple times and it keeps getting it wrong. When it does get it right, if you ask why it keeps getting it wrong it will then “correct” itself back to the wrong answer…. It even claims to be the smartest AI available According to Grok many AI models get 17 wrong. It has to do with LLMs. I asked Grok and it gets it right no problem It’s really wild that we are putting so much faith in AI when currently there are issues like this on such simple things

This is wild ChatGPT 6 is asked how many of the letter “E” are in the word seventeen. It can’t get the correct answer no matter how many times you ask. Even if you make it count out the letters, it still can’t get it right It will actually get an attitude and show frustration with you for trying to get the model to say the correct answer I ran the test multiple times and it keeps getting it wrong. When it does get it right, if you ask why it keeps getting it wrong it will then “correct” itself back to the wrong answer…. It even claims to be the smartest AI available According to Grok many AI models get 17 wrong. It has to do with LLMs. I asked Grok and it gets it right no problem It’s really wild that we are putting so much faith in AI when currently there are issues like this on such simple things

Wall Street Apes

90,294 views • 17 days ago

Meta released SAM3 a model to reconstruct objects from segmented images. I tried it and here are 4 things I found out - The generation is incredibly fast when compared with other 3D GenAI models - The models generated are not meshes but Gaussian Splatting. I was surprised that in the paper it was not mentioned anywhere - When generating multiple objects the arrangement is a bit messed up. - Unfortunately, it is impossible to download any of the .ply generated 😔 I cannot stop thinking about the implications of this in XR. Objects in the real world could be almost instantly turned into photorealistic copies and even rearranged to the user's wish.

Meta released SAM3 a model to reconstruct objects from segmented images. I tried it and here are 4 things I found out - The generation is incredibly fast when compared with other 3D GenAI models - The models generated are not meshes but Gaussian Splatting. I was surprised that in the paper it was not mentioned anywhere - When generating multiple objects the arrangement is a bit messed up. - Unfortunately, it is impossible to download any of the .ply generated 😔 I cannot stop thinking about the implications of this in XR. Objects in the real world could be almost instantly turned into photorealistic copies and even rearranged to the user's wish.

Gabriele Romagnoli

32,447 views • 10 months ago

Astra built this three.js game with gorgeous graphics, then recreated every scene in Blender and made the trailer too. It feels like a massive upgrade from Sol for 3D graphics and front-end design. The UI feels way more polished and less like AI slop. The rocks, trees, monuments, and characters are all procedural code, keeping the game under 2 MB before the image and music assets. The level of autonomy is unbelievable. It needed very little steering, though I still had to keep pushing it to work harder on the details. The UI icons, buttons, and ornaments were made as transparent PNGs with GPT Image 2.

Astra built this three.js game with gorgeous graphics, then recreated every scene in Blender and made the trailer too. It feels like a massive upgrade from Sol for 3D graphics and front-end design. The UI feels way more polished and less like AI slop. The rocks, trees, monuments, and characters are all procedural code, keeping the game under 2 MB before the image and music assets. The level of autonomy is unbelievable. It needed very little steering, though I still had to keep pushing it to work harder on the details. The UI icons, buttons, and ornaments were made as transparent PNGs with GPT Image 2.

Meng To

137,306 views • 28 days ago

Cover evolution for Maintenance: Of Everything, Part One. “The cover of the book honors the idea of Kintsugi, the Japanese art of basically repairing broken pottery with a kind of a gold glue. And so it not only fixes it, but it makes it more beautiful. And you honor the mistake that broke it, and you honor the repair . . . So Kintsugi is a way of kind of just honoring the fact that things do break. But nobody actually wants things to break.” —Stewart Brand Design Pablo Delcan, Art direction 🎨

🎨

216,182 views • 8 months ago

Took Grok 4.6 out for a spin this weekend and built this horde shooter prototype on ThreeJS. It is a big step up from Grok 4.5! I've always wanted to try building a horde shooter game with thousands of enemies onscreen and used Grok 4.6 for pretty much everything (including UI, vfx, and even the sfx!), except the 3D models for the characters, which were made and rigged with Meshy. I'm also amazed that ThreeJS with WebGPU can really handle thousands of enemies onscreen at the same time, though I think I pushed the limit here with Depth of Field and some other fancy shaders. Very fun!

Took Grok 4.6 out for a spin this weekend and built this horde shooter prototype on ThreeJS. It is a big step up from Grok 4.5! I've always wanted to try building a horde shooter game with thousands of enemies onscreen and used Grok 4.6 for pretty much everything (including UI, vfx, and even the sfx!), except the 3D models for the characters, which were made and rigged with Meshy. I'm also amazed that ThreeJS with WebGPU can really handle thousands of enemies onscreen at the same time, though I think I pushed the limit here with Depth of Field and some other fancy shaders. Very fun!

Danny Limanseta

65,105 views • 1 month ago

The interesting part about GPT in CAD is not that it can make a shape. It is that through MCP, GPT can work directly inside Onshape and turn a plain-English description into actual CAD geometry you can keep editing. In this case, the model builds a radial mechanical part with a central hub, repeated ribs, an outer ring, and a full pattern of mounting holes, all inside the real Onshape document. There is no fake “3D-looking” render hiding the hard part. The geometry exists in the CAD workspace, can be rotated, inspected, modified, and used as the starting point for the next iteration. That changes the workflow from manually clicking through every feature to describing the structure, letting GPT build the first pass, then correcting what matters. Once MCP gets reliable enough, the boring part of CAD starts looking less like modeling and more like supervising.

The interesting part about GPT in CAD is not that it can make a shape. It is that through MCP, GPT can work directly inside Onshape and turn a plain-English description into actual CAD geometry you can keep editing. In this case, the model builds a radial mechanical part with a central hub, repeated ribs, an outer ring, and a full pattern of mounting holes, all inside the real Onshape document. There is no fake “3D-looking” render hiding the hard part. The geometry exists in the CAD workspace, can be rotated, inspected, modified, and used as the starting point for the next iteration. That changes the workflow from manually clicking through every feature to describing the structure, letting GPT build the first pass, then correcting what matters. Once MCP gets reliable enough, the boring part of CAD starts looking less like modeling and more like supervising.

Rina

16,772 views • 1 month ago

Blended-NeRF: Zero-Shot Object Generation and Blending in Existing Neural Radiance Fields paper page: Editing a local region or a specific object in a 3D scene represented by a NeRF is challenging, mainly due to the implicit nature of the scene representation. Consistently blending a new realistic object into the scene adds an additional level of difficulty. We present Blended-NeRF, a robust and flexible framework for editing a specific region of interest in an existing NeRF scene, based on text prompts or image patches, along with a 3D ROI box. Our method leverages a pretrained language-image model to steer the synthesis towards a user-provided text prompt or image patch, along with a 3D MLP model initialized on an existing NeRF scene to generate the object and blend it into a specified region in the original scene. We allow local editing by localizing a 3D ROI box in the input scene, and seamlessly blend the content synthesized inside the ROI with the existing scene using a novel volumetric blending technique. To obtain natural looking and view-consistent results, we leverage existing and new geometric priors and 3D augmentations for improving the visual fidelity of the final result. We test our framework both qualitatively and quantitatively on a variety of real 3D scenes and text prompts, demonstrating realistic multi-view consistent results with much flexibility and diversity compared to the baselines. Finally, we show the applicability of our framework for several 3D editing applications, including adding new objects to a scene, removing/replacing/altering existing objects, and texture conversion.

Blended-NeRF: Zero-Shot Object Generation and Blending in Existing Neural Radiance Fields paper page: Editing a local region or a specific object in a 3D scene represented by a NeRF is challenging, mainly due to the implicit nature of the scene representation. Consistently blending a new realistic object into the scene adds an additional level of difficulty. We present Blended-NeRF, a robust and flexible framework for editing a specific region of interest in an existing NeRF scene, based on text prompts or image patches, along with a 3D ROI box. Our method leverages a pretrained language-image model to steer the synthesis towards a user-provided text prompt or image patch, along with a 3D MLP model initialized on an existing NeRF scene to generate the object and blend it into a specified region in the original scene. We allow local editing by localizing a 3D ROI box in the input scene, and seamlessly blend the content synthesized inside the ROI with the existing scene using a novel volumetric blending technique. To obtain natural looking and view-consistent results, we leverage existing and new geometric priors and 3D augmentations for improving the visual fidelity of the final result. We test our framework both qualitatively and quantitatively on a variety of real 3D scenes and text prompts, demonstrating realistic multi-view consistent results with much flexibility and diversity compared to the baselines. Finally, we show the applicability of our framework for several 3D editing applications, including adding new objects to a scene, removing/replacing/altering existing objects, and texture conversion.

AK

62,768 views • 3 years ago