AI · Web app
TRELLIS
AI model that generates textured 3D assets from single images
TRELLIS is an AI-powered 3D generation tool available as an open web demo on Hugging Face Spaces. It transforms single 2D images into detailed 3D models with normal maps and allows users to export them directly as GLB files. It provides a fast, accessible way to test state-of-the-art image-to-3D generation directly in the browser.
What it's for
- Turn 2D concept art and illustrations into textured 3D assets
- Rapidly prototype 3D models for games and VR environments
- Extract GLB files from single reference photos for 3D workflows
Features
- Single-image and multi-view 3D model generation
- Automatic background removal via rembg or built-in alpha channel detection
- Interactive 3D and normal map preview directly in the browser
- Export and download assets directly to GLB format
- Runs online using Hugging Face ZeroGPU
Pricing
Free to use on Hugging Face Spaces
Details
- Type
- Web app
- Category
- AI
- Topic
- 3D Generation
- Open source
- Yes
- Platforms
- Web
- For
- Game developers, 3D artists, indie creators, and AI researchers
- Alternative to
- Tripo3D, Meshy, Luma AI
- Tags
- 3d, image to 3d, glb, zero gpu, huggingface, generative ai
Why it was saved
- @victormustar: Image-to-3D generation in one step: upload a PNG (background removed with rembg if no alpha channel), click Generate to get a textured 3D model plus normal map, then Extract GLB and Download GLB to get a production-usable 3D file. The video shows a 2D mushroom-house illustration becoming a clean 3D model, supporting the claim that 3D generation is nearly production-ready.
Compare
Related
Turn text prompts and 2D images into textured, printable 3D models.
Turn freehand sketches into textured 3D models inside a Three.js scene
Rebuilds reference images as procedural, animation-ready Three.js models in code.
Open-source 4B flow-matching model generating textured 3D assets from single images.
AI generator converting text and 2D images into 3D models.
Spatial AI models that turn images, text, and video into 3D worlds
Updated