How to Install an AI 3D Model Generator on Your Computer for Free: A Step-by-Step Guide

From text prompts or 2D images, create detailed 3D models with textures. Here’s a step-by-step guide on installing and using this AI model on your computer.
github.com/Yan...
visualstudio.m...
developer.nvid...
My recommendations:
Must have VPN: go.nordvpn.net...
Topaz AI Video and Photo processing: topazlabs.com/...
Stuff I am using:
www.amazon.com...
My Vue book - amzn.to/2TGUkvQ
3D Art essentials - amzn.to/2RfqPjh
My Patreon webpage - / geekatplay
Tutorials and packs - gumroad.com/ge...
Tutorials website - www.geekatplay...
Photography - www.chopinepho...
Subscribe to my channel for fast notifications on new tutorials - / @geekatplay

[@Geekatplay]

Check updated video: kzbin.info/www/bejne/gWbVhoyMiN-koc0

[@kenmorris2858]

Another excellent tutorial, many thanks....

[@Geekatplay]

You’re welcome! Glad you found it helpful.

[@karlisstigis]

Thank you!

[@Geekatplay]

You're welcome!

[@Simansh]

Очень приятно вас слушать и слышать) Привет из Ленинграда!)

[@Geekatplay]

Спасибо большое за тёплые слова! Очень приятно получить привет из Ленинграда, передаю ответный привет!

[@computervision557]

A beginner of blender and 3D modeling, the results are stunning, but the "lines“ on 13:29 are extremely complicated I guess these model could be very hard to finetune manually by blender?

[@Geekatplay]

Same ai model now has plugin in blender

[@louiewashere3]

I get this error: ERROR: [Errno 10048] error while attempting to bind on address ('0.0.0.0', 8080): [winerror 10048] only one usage of each socket address (protocol/network address/port) is normally permitted

[@Geekatplay]

This error means that another application or process is already using port 8080. Here’s how to fix it: Check for Running Applications: Look for any applications or servers already using port 8080. You can change their port or stop them temporarily. Find and Stop the Process: Run this command in Command Prompt to identify the process using the port: netstat -ano | findstr :8080 Note the PID (Process ID) from the result. Then, stop the process by running: taskkill /PID /F Change the Port for Your Application: If you don’t want to stop the other process, you can run your server on a different port. For example, if you are starting an API server, specify a different port: python api_server.py --host 0.0.0.0 --port 8081 Let me know if this resolves the issue or if you need more help!

[@wcysitp]

it's always 40k faces. is it possible to control the number of faces in the result?

[@Geekatplay]

not as i saw, may be in config file, i do wish it was mesh adjustment options.

[@trigonumvr5129]

Remember to restart Hunyuan3D-2 every time you change the ‘max_facenum:’ value, otherwise the change to the ‘max_facenum:’ variable will not be executed

[@UnderPantsGG]

When opening 1 folder, it gives the error ERROR: Failed to build installable wheels for some pyproject.toml based projects (custom_rasterizer). How to solve it, or is this how it should be? writes: This error originates from a subprocess, and is likely not a problem with pip.

[@Geekatplay]

This error usually happens if some dependencies or tools are missing. First, update pip, setuptools, and wheel with pip install --upgrade pip setuptools wheel. Next, ensure you have a C++ compiler installed, like Visual Studio Build Tools. Then, retry the installation with python setup.py install. If it still doesn’t work, check if there’s a prebuilt version of the library or consult the project’s documentation for help. Let me know if you need more assistance!

[@UnderPantsGG]

@@Geekatplay I downloaded everything, updated it, but this error still remains

[@Geekatplay]

I had a similar problem and ended up removing/uninstalling almost all tools (PyTorch, CUDA, etc.) and reinstalling them. Make sure that CUDA and PyTorch are compatible with the same version. In my case, I believe it was version 1.18.

[@midnightstories5695]

OK a little confused when updating to the 2nd version with Generating textures do you overwrite the first version folder or do you keep then in separate folders ? thanks

[@Geekatplay]

What do you mean by “second version”? If you’re referring to the installation, I usually create a new setup for each test, but I test them separately and keep the one that works best for me. If you mean installing the model generator first and then textures, etc., just leave it as is. If you’re talking about model generation, the app creates unique folders for each newly generated model.

[@midnightstories5695]

@@Geekatplay I installed the shapes one first, then there's a different zip for the textured one so you just do them in separate folders not in the same.

[@remedyrenders5746]

Yah for me also the generation of textures is blocked I can only create the shapes

[@Geekatplay]

be sure if shape working, then install step 'one'. As well stop and restart server.

[@kobayashi19891805]

I installed hunyuan 3d on pinokio but it doesn't let me use texture :(

[@AdamD222]

how did you do that ,i find problems installing it and is it possible to install it with Intel Iris Xe Graphics (integrated cpu )insteat of NVIDIA CPU?

[@Geekatplay]

I haven’t personally tested it on Pinokio, so it’s hard to say what might be wrong. You could follow my installation process-though when I installed Hunyuan 3D on Pinokio, it wouldn’t let me use textures either

[@dadadies]

@@AdamD222 Intergrated Intel GPU has like 1 to 2 GB VRAM. You need at lest 6GB VRAM and it usually has to be Nvidia GPU.

[@barrymcbess4190]

Leider bekomme ich das bei mir nicht zum laufen. Da ich leider kein englisch spreche und mit der Deutschen Übersetzung arbeiten muss ist es leider etwas konfus. Einfach mal die Reihenfolge wann man was installieren muss aufschreiben. Trotzdem danke für deine Mühe!!

[@Geekatplay]

Schade, dass es bei dir nicht funktioniert. Ich verstehe, dass es mit der Übersetzung etwas verwirrend sein kann. Am besten geht man Schritt für Schritt vor und installiert zuerst die benötigten Programme (Python, PyTorch), dann die Modelle und zum Schluss die benötigten Erweiterungen (xFormers oder andere). Ich werde versuchen, in Zukunft eine klarere Reihenfolge in meinen Videos oder in einer kurzen Anleitung zusammenzufassen. Vielen Dank für dein Feedback und deine Geduld!

[@barrymcbess4190]

@@Geekatplay Danke für deine Antwort. Ich habe das Programm in dieser Reihenfolge installiert. 1.) download -Models 2:) Cuda-12.4 3.) Visual Studio das ist alles.

[@maxlopez12123]

Brilliant, I was wondering how you install this thing

[@Geekatplay]

I'm glad you found the installation process helpful! 👍

[@cblack5071]

is there a google collab for this?

[@Geekatplay]

not as i know, but on their website they have links to the test servers

[@LouisGedo]

👋 hi 👋

[@Geekatplay]

Hi, thank you for your support!!!!

[@MaiaCoimbra]

very poor yet

[@Geekatplay]

give some time, first ai images was horrible

[@himanshusingh4451]

is RTX 4060 8 gb will be good for this ??

[@Geekatplay]

yes, it will works great