# Stability-AI/stable-point-aware-3d

SPAR3D: Stable Point-Aware Reconstruction of 3D Objects from Single Images

Repository: https://github.com/Stability-AI/stable-point-aware-3d
Canonical: https://ross.abutalabs.com/products/stable-point-aware-3d
Homepage: https://spar3d.github.io/
Language: Python
License: NOASSERTION
License Family: other
Last push: 2025-05-05T16:44:13+00:00

## Health v2 (maintenance only)
Score: 30/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 20, release rhythm 35, longevity 45
- inputs: {"age_days": 638, "days_push": 485, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1068, forks 108 (observed 2026-08-28T04:03:27.568948+00:00)

## What it is
SPAR3D is Stability AI's open-source model for fast single-image 3D mesh reconstruction using a two-stage pipeline with point cloud conditioning. It includes a point diffusion model, editable intermediate point clouds, and a Gradio demo, achieving ~0.7 second inference.

## Use cases
- reconstruct a 3D mesh from a single photo
- generate 3D models from images for games or AR
- edit the backside of a reconstructed mesh via point cloud
- fast image-to-3D asset creation pipeline
- run single-image 3D reconstruction on GPU locally

## When to choose
- you need fast (sub-second) single-image to 3D mesh conversion
- you want editable intermediate point clouds to fix occluded geometry
- you need a state-of-the-art open-source image-to-3D model with a demo

## When to avoid
- you need multi-view or video-based 3D reconstruction
- you have no GPU and cannot tolerate slow low-VRAM mode
- you need permissively licensed code for commercial use without checking the custom Stability AI license

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, image-processing, graphics, simulation
- domain: computer-vision, graphics, artificial-intelligence, deep-learning
- platform: python, windows
- tags: 3d-reconstruction, single-image-to-3d, mesh-generation, point-cloud, diffusion-model, stability-ai, spar3d, gradio-demo, gpu, linux, macos

## Member repositories
- Stability-AI/stable-point-aware-3d (main) score 30

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:27.568948+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 2026-08-30T06:54:41.176243+00:00, confidence not recorded.
  - readme: https://github.com/Stability-AI/stable-point-aware-3d (fetched 2026-08-28T04:03:27.568948+00:00, sha 242f8d361618)
  - homepage: https://spar3d.github.io/ (fetched 2026-08-29T12:56:58.806902+00:00, sha 5d2f25578ca0)
- Data as of 2026-08-30T08:39:29.467469+00:00.