I started using Blender through MCP about two weeks ago, and I quickly realized that you can build almost anything with AI. This model was created using Blender, Hunyuan3D, Gemini, and ChatGPT. Here’s how I did it: I opened Gemini, uploaded an image of the Gundam model, and asked... it to generate a clean front-view image. I uploaded that front view to ChatGPT and asked it to generate two additional angles: a back view and a 45-degree front-left view. I went to: I signed up with my email and translated the page into English. Then I opened Image to 3D and selected the multi-image option. You’ll see a diagram of a whale from several angles. Upload each reference image in its corresponding position, such as front, 45-degree front-left, and back. I selected the 1.5M-face option. This produces a very high-poly model, but don’t worry, we’ll fix that next. Once the generation is complete, download the model as a GLB file. From the Hunyuan homepage, open 3D Studio using one of the dropdown menus. Select the topology or retopology tool and upload your GLB. I chose the High setting to preserve as much detail as possible. After a few seconds, the model was retopologized. It kept most of its visual detail while using far fewer polygons. The original head and rifle didn’t look very good, so I generated them separately. I returned to ChatGPT and created dedicated reference images for the Gundam’s head and rifle. I generated each part individually in Hunyuan at the 1.5M setting, then ran both through the same retopology process. Next came the textures. Open the texture section, select the multi-image option, and upload the same reference images according to the whale orientation indicators. However, instead of using the original clean textures, I asked ChatGPT to recreate them with wear, rust stains, scratches, and other surface damage. This gave the Gundam a much older and more authentic appearance. Once every part was textured, I imported everything into Blender. You can ask Codex through MCP to remove the original head and rifle, or you can do it manually. Select the main model and press Tab to enter Edit Mode. Press 3 to enable face selection, then press C to activate Circle Select. Paint over the faces belonging to the original helmet or rifle. You can press X to delete those faces or P to separate them into another object. Then position the newly generated, more detailed head and rifle in their place. I demonstrate this process in one of my older videos. And that’s it. You now have a very cool 3D Gundam model! Afterward, I created the cockpit, separated the model into movable sections, and rigged everything for use in my Three.js game. That process deserves its own tutorial, though. If anyone wants to see it, let me know. Or just ask your AI, I guess. They seem to know everything these days. xDshow more

testing gameplay

That's insane between 1999 and 2004, I worked on a Dragon Ball Z Quake 3 Total Conversion and a Gundam Quake 3 Total Conversion and this brings back memories of the models we pushed at the time. We were pushing around 5000 polygons and growing which at the time was quite high for 3D engines!

Looks nice! I also made mech like yours in my game using blender mcp

that was amazing! love the retro graphics

I think @IsThisA3DModel is going to have a bad time with this

The topology and textures are pure shit. Congrats.

Need this in my life but I'm already doing so much fuck

Dont worry, im making a video tutorial, you can just feed the tutorial into your coding Ai and i believe he will be able to do it all autonomously lol

I’d love to see that tutorial

You should also try out @flickartHQ ‘s new 3D Stage, blender like creation but the agent creates for you in seconds.

can i see the uv?

Now the worst part, weight painted, bones, animation.

Now the rogging and three.js tutorial pls

Chatgpt did the rigging for me lol

Wow

How much is the tencent service? Is it the same as

it was free in ten ent

this is pretty cool. Would love to see it!

Neat you can do that. But every game dev standard, that model is just horrendous. A lot of unnecessary geometry in places resulting to: Not subdiv ready. Shading errors Hard surface modeling nightmare. Bad optimization

The model is already low poly, made with polygen, its running on 60fps on a browser, with shadows, vfx etc

Low poly hides the issues and constrains the model to a single purpose. Shield, chest vents & circles on shoulders have so much geometry. N-gons on many parts of the model.

i have a question, i don't know much about 3d unfortunately but i'm trying to have ai create a simple orbit swish in retract and orbit swish out motion on a can product commercial that i'm trying to make, i genuinely do not know how to drive the blender mcp to achieve this.

like, product animation? i think you can tell mcp to set the origin of the camera to the can, or set thje camera itself as parent, so you can animate the orbit motion and then animation the camera closing in on an axis from the camera to the parent object, although i never done anything like this so im just guessing

😇😇😇😇感觉好麻烦

A simpler workflow would be nicer but it looks cool

Thanks for open sourcing your workflows

Sick

Yes, it’s great … I can use ChatGPT to help write a storyline, then generate concept art for characters by reading the storyline and then generate multiple views of characters than can be used by a 3d model generator like Meshy AI that can be edited by ChatGPT over the blender or Unity3d MCP interfaces.

yup, essentially a ford style production pipeline

Yeah, and it’s not a replacement for human artistry or talent as many people think - AI augments those things. I’ll be honest, I’ve been using Blender for about 8 years now and I LOVE not having to spend hours, days and weeks crafting 3d models from scratch anymore and to be able to go from 2d images to 3d models with ease! Reworking topology and textures sometimes, is much better than starting from scratch and the MCP interfaces are wonderful for a lot of things like repetitive tasks!😃💙👍

I started using belnder because of mcp and often i find myself doing stuff manually rather than using ai because its so satisfying

Yeah, I needed to simplify the geometry of a building done by a talented environment artist so I asked for a blend file containing planes with rendered from all sides applied to planes after switching all the textures to emissive. The original building mesh had lots of meshes for individual boards, so the vertex count was very high. ChatGPT did it all in one shot in just a few minutes. 😅🤣 But “cutting out” the walls and adding back normal maps, from the generated planes having the new textures, was a lot of fun. I know what you mean! 😃

this is cheating at 3d

pretty much, lol. i just spawned another 3 mechs in the blink of an eye too, right now separating the limbs to animate and then install the cockpit in them, i could make them super high detail too, but since im using threejs the low poly is fine

Great process. Thanks for sharing. “Ai enhanced” workflows are the hardest to find on here it seems. It’s either “I one-shotted this game with AI blah” or “watch me spend 300 hours of my life modeling this detail” haha 😆

Looks cool. You can take it to the next level and import this model and use it as an FBX/GLB inside Creadev ( which is a no-code platform for developing AA/Triple-i quality games, animations and simulations.

First candidate: Using Gemini to upload a Gundam model demonstrates how AI can transform simple images into complex 3D models.

Thanks for the explanation, but you skipped the rig and the animation parts. Why don't you share your process for doing so too?

Been using this workflow for a while now. The latest 3.1 model is very good. I’m still shocked we get 20 free generations per day for free at this level of quality.

Looks good

Am using chatgbt to generate frames for characteristics for 2d animation. Looking for interested fellows to do projects

WTF is that topology? Jesus fucking Christ that is bad.

Absolute cinema! I am experimenting with unity cli, codex, Gemini 3.7 flash and ChatGpt, made a playable 5min 2d space shooter level in around 8h total

does it all for you

the hard part is mesh cleanup after hunyuan3d not the first model generation

Can you share which Blender MCP and how you use?