Skip to content
SPECIFICTION
Get listed

Wonder3D

github.com

WebWindowsLinuxSelf-hosted
Visit github.com
Spec score 7.1 #23 of 24 in AI 3D Model Generators

Spec sheet

Purpose
Wonder3D reconstructs textured 3D meshes from a single-view image.Read 7 Oct 2026 · github.com
Platforms
The README provides Linux setup, points Windows users to a separate branch, and links a Hugging Face demo.Read 7 Oct 2026 · github.com
Integrations
The documented workflow uses Hugging Face Diffusers and checkpoints, and describes using SAM and Clipdrop or rembg for foreground masks.Read 7 Oct 2026 · github.com
License
The repository states that Wonder3D is under the MIT license and free to use with acknowledgement.Read 7 Oct 2026 · github.com
Founded
2023Read 28 Sep 2026 · github.com
Contact and collaboration
The project invites collaboration and lists [email protected] for contact.Read 7 Oct 2026 · github.com
Generation
It generates consistent multi-view normal maps and corresponding color images using cross-domain diffusion.Read 7 Oct 2026 · github.com
Generation method
It generates consistent multi-view normal maps and corresponding color images with a cross-domain diffusion model, then uses normal fusion for reconstruction.Read 7 Oct 2026 · github.com
Hardware
The example pipeline moves to CUDA when available, and the repository provides a one-GPU inference configuration.Read 7 Oct 2026 · github.com
Input guidance
The README says front-facing images with fewer occlusions generally produce better reconstructions.Read 7 Oct 2026 · github.com
Input preparation
The README suggests using Clipdrop or rembg to segment the foreground object.Read 7 Oct 2026 · github.com
Intended uses
The project page shows examples involving Houdini, 3D asset creation, 3D printing, 2D designs, and animal images.Read 7 Oct 2026 · xxlong.site
Local demos
The repository provides local Gradio demos for generating normals and colors and for mesh reconstruction.Read 7 Oct 2026 · github.com
Mesh extraction
It provides Instant-NSR and NeuS-based mesh extraction workflows.Read 7 Oct 2026 · github.com
Model dependencies
The inference instructions call for downloaded checkpoints and a Segment Anything Model checkpoint.Read 7 Oct 2026 · github.com
Multi-view attention
The method uses multi-view cross-domain attention to exchange information across views and image modalities.Read 7 Oct 2026 · xxlong.site
Notable limits
The README says the implementation supports six views and 256×256 generation, and resizes input images to that resolution.Read 7 Oct 2026 · github.com
Output limits
The current implementation supports six views and resizes input images to 256×256 for generation.Read 7 Oct 2026 · github.com
Processing time
The project says reconstruction takes about 2–3 minutes.Read 7 Oct 2026 · xxlong.site
Reconstruction caveat
The README says occluded inputs can produce worse reconstructions because six views may not cover the complete object.Read 7 Oct 2026 · github.com
Reconstruction options
The repository documents Instant-NSR and NeuS mesh extraction; it says NeuS uses less GPU memory and favors smooth surfaces but takes longer and may produce less sharp textures.Read 7 Oct 2026 · github.com
Reconstruction speed
The project says reconstruction takes about 2–3 minutes.Read 7 Oct 2026 · xxlong.site
Support and collaboration
The README invites questions and collaboration by email at [email protected].Read 7 Oct 2026 · github.com
Supported systems
The README provides Linux setup instructions and points Windows users to a separate setup branch.Read 7 Oct 2026 · github.com
Training
The repository includes two training scripts and describes a two-stage training process.Read 7 Oct 2026 · github.com

Where Specifiction places it

Alternatives to Wonder3DAll alternatives →