Furkan Gözükara
patreon
The Very Best Open Source Image to 3D APP TRELLIS Installers Updated
🕑 Added 2025-01-16 00:28:09 +0000 UTCTRELLIS is published by Microsoft and it is open source state of the art image to 3D APP. Check above images to see what it can make
It can generate high quality GLB file and also Gaussian Splatting ply file that you can use in your favorite apps.
A GLB file (. glb), which stands for “GL Transmission Format Binary file”, is a standardized file format used to share 3D data.
Installers published here : https://www.patreon.com/posts/117470976
It is really lightweight and we have installers for Windows, RunPod and Massed Compute.
Our installer generates a Python 3.10 VENV and install everything into it.
With our latest update now it is even easier.
Installers published here : https://www.patreon.com/posts/117470976
So now we Windows installer installs pre-compiled diffoctreerast and diff_gaussian_rasterization
You can edit Windows_Install.bat and still compile on your system
So who were having CUDA and C++ mismatch compiling errors, this will fix the issue for you hopefully
Also newest version now supports multiple image upload from different angles of the object / subject
More info about TRELLIS
TRELLIS Official REPO : https://github.com/microsoft/TRELLIS
TRELLIS is Structured 3D Latents for Scalable and Versatile 3D Generation
TRELLIS is a large 3D asset generation model. It takes in text or image prompts and generates high-quality 3D assets in various formats, such as Radiance Fields, 3D Gaussians, and meshes. The cornerstone of TRELLIS is a unified Structured LATent (SLAT) representation that allows decoding to different output formats and Rectified Flow Transformers tailored for SLAT as the powerful backbones. We provide large-scale pre-trained models with up to 2 billion parameters on a large 3D asset dataset of 500K diverse objects. TRELLIS significantly surpasses existing methods, including recent ones at similar scales, and showcases flexible output format selection and local 3D editing capabilities which were not offered by previous models.
🌟 Features
High Quality: It produces diverse 3D assets at high quality with intricate shape and texture details.
Versatility: It takes text or image prompts and can generate various final 3D representations including but not limited to Radiance Fields, 3D Gaussians, and meshes, accommodating diverse downstream requirements.
Flexible Editing: It allows for easy editings of generated 3D assets, such as generating variants of the same object or local editing of the 3D asset.
Comments
Furkan Gözükara
sadly i dont know at the moment if any scripts supports :(
goose1969x
I was looking to train a model off of Tripo SR but saw that they do not have any training scripts. Do you know if Trellis will allow for custom model training?
Furkan Gözükara
no i didnt try comfyui. you know generating images easy you can use any ai model :D
The AI Diva
Fair enough! Have you tried the comfyui version of this? There’s also a Gradio version but it’s missing text to image…
Furkan Gözükara
well i need more information. record a short video and send me so i can analyze. also what gpu do you have?
Furkan Gözükara
well i need more information. record a short video and send me so i can analyze. also what gpu do you have?
Anshul Gupta
I AM not able to swap face of A video
Anshul Gupta
I AM unable to swap face in A video
Furkan Gözükara
ignore it, it is looking camera and cant find. also it is not my error but the developer of Alucard :)
Anshul Gupta
Your one-click installer for Rope_Alucard24_V13 is not working properly; it is throwing the following error in processing. "[ERROR:0@14.908] global obsensor_uvc_stream_channel.cpp:158 cv::obsensor::getStreamChannelGroup Camera index out of range"
Furkan Gözükara
sadly they didnt add. making docker is really hard but this installer is pretty optimized and quickly completed on massed compute atm. only initial starting is taking time like 15 min atm. i hope they will add permanent storage soon
The AI Diva
Looks amazing. Has massed compute added persistent storage?? Would love to see you add docker files for these builds so we can use our own servers
