# xxlong0/Wonder3D

Single Image to 3D using Cross-Domain Diffusion for 3D Generation

Repository: https://github.com/xxlong0/Wonder3D
Canonical: https://ross.abutalabs.com/products/wonder3d
Homepage: https://www.xxlong.site/Wonder3D/
Language: Python
License: MIT
License Family: permissive
Topics: 3d-generation, 3d-aigc, single-image-to-3d, 3dgeneration
Last push: 2025-03-14T07:09:19+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 11, release rhythm 35, longevity 75
- inputs: {"age_days": 1054, "days_push": 537, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 5425, forks 439 (observed 2026-08-28T04:09:17.957693+00:00)

## What it is
Wonder3D is a cross-domain diffusion model that reconstructs high-fidelity textured 3D meshes from a single image in 2-3 minutes. It generates consistent multi-view normal maps and color images, then fuses them via a geometry-aware normal fusion algorithm.

## Use cases
- convert a single photo into a 3D model
- generate textured mesh from one image
- reconstruct 3D geometry from a single-view image
- create multi-view normal maps from an image
- fast image-to-3D asset generation for games
- 3D content creation from photos

## When to choose
- you need fast (2-3 minute) single-image to 3D reconstruction
- you want high-quality textured meshes with detailed geometry
- you need a research-grade image-to-3D pipeline with a Hugging Face demo
- you want multi-view consistency without slow per-shape SDS optimization

## When to avoid
- you need production-grade 3D generation at scale
- you require editable, artist-ready topology rather than fused meshes
- you cannot run GPU inference locally
- you need a native 3D diffusion model rather than a 2D multi-view approach

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, image-processing, graphics, stable-diffusion
- domain: computer-vision, graphics, artificial-intelligence, deep-learning
- platform: python, cross-platform
- tags: 3d-generation, image-to-3d, diffusion-model, multi-view, mesh-reconstruction, normal-maps, cvpr-2024, textured-mesh, gpu, linux

## Member repositories
- xxlong0/Wonder3D (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:17.957693+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:56:55.520533+00:00, confidence not recorded.
  - readme: https://github.com/xxlong0/Wonder3D (fetched 2026-08-28T04:09:17.957693+00:00, sha fb344a90563f)
  - homepage: https://www.xxlong.site/Wonder3D/ (fetched 2026-08-29T08:52:10.933132+00:00, sha 65b0710d6a64)
- Data as of 2026-08-30T08:39:29.467469+00:00.