LGM (Large Gaussian Model)
High-resolution 3D Gaussian generation from text or single images.
About
LGM, the Large Multi-View Gaussian Model, generates high-resolution 3D Gaussian Splatting representations from a text prompt or a single image, producing assets in seconds using an asymmetric U-Net backbone. It pairs with MVDream and ImageDream for the multi-view stage and needs roughly 10 GB of GPU memory for inference. Developed by 3DTopia. Released under the MIT license.
Reviews (0)
Leave a Review
No reviews yet. Be the first to review!
Details
- Category
- 3D Model Generation
- Price
- Free
- Platform
- Local/Desktop
- Difficulty
- Intermediate (3/5)
- License
- MIT
- Minimum VRAM
- 8 GB
- Added
- Apr 3, 2026
Related Tools
Novel view synthesis from a single image using diffusion model conditioning.
Text-to-3D generation using Gaussian Splatting with multi-view guidance.
Fast single-image 3D reconstruction by Stability AI and Tripo in under 1 second.
Fast 3D content generation using Gaussian Splatting with mesh extraction.
Large-scale 3D asset generation framework using structured latent representations.
Open-source 3D reconstruction pipeline based on photogrammetry.