Hugging Face's logo

Join the conversation

Join the community of Machine Learners and AI enthusiasts.

Sign Up
Back to feed
aiqtech 
posted an update 16 days ago
Post
2881
🎨 SORA 3D: Create 3D Models from Text and Images
Hey there! Today I'm excited to share 'SORA 3D', a project that generates 3D models from text prompts or images.
✨ Key Features

3D generation from text/image input
Multilingual prompt support
Automatic GLB conversion
Real-time 3D preview
Mesh optimization & texture quality control

🚀 How to Use

Enter text or upload image
Adjust generation settings (optional)
Click 'Generate 3D'
Extract and download GLB file

🛠 Tech Stack

Hugging Face Transformers
PyTorch
Gradio
TRELLIS image-to-3D conversion
FLUX image generation

💡 Use Cases

Game asset creation
Metaverse content
Product prototyping
Educational 3D models

LINK: ginipick/SORA-3D

Try the demo and let me know what you think! 😊
#AI #3DGeneration #MachineLearning #HuggingFace #ComputerVision
usmanayaz

But its access is limited to few countries initially I guess

·
YaTharThShaRma999

This doesn’t even use sora, it uses TRELLIS which is an open source 3d model by Microsoft under mit license and does not limit access to a few countries.

chrismorin

This is awesome, what do we use for the APP environment variable if we want to duplicate?

Mahasti

Hi, I am very new to Treliss. I tried to convert images (.PNG, .JPEG) to 3D model. I followed the process and Extract and downloaded GLB that took me to Adobe Substance 3D Stager. However, it didn't download the full image there. Any suggestions?

Also, any tutorial on how touse The TRELLIS? Thanks so much!

In this post