CPU3DAI tools for 3D, video and audio

Stable Fast 3D: 3D Generator — What It Does and What You Need to Run It

Stable Fast 3D is an open source 3D generator from Stability AI that solves the image-to-3D task. The tool reconstructs a 3D mesh from one or more images, performing UV unwrapping and separating lighting from geometry.

What it does

  • Reconstruct a 3D model from one or more images (image-to-3D).
  • Perform UV unwrapping and Illumination Disentanglement — as described by the author.
  • Save the result in GLB format.
The authors do not list any other export formats.

What you need to run it

  • Platform: CUDA or MPS (as described by the author).
  • VRAM: about 6 GB for a single image (as described by the author).
  • System RAM: the authors recommend running the CPU version if your system has less than 32 GB of unified memory — this is the developer's claim, not a measurement.
  • Language: Python.
  • Weight files: 3.7 GB (one file).
  • Code license: Stability AI Community License: free commercial use up to 1 million dollars in revenue.
  • Weights license: Stability AI Community License: free up to $1 million in revenue.
The code is available in the repository at github.com/Stability-AI/stable-fast-3d, the model weights at huggingface.co/stabilityai/stable-fast-3d. Project website: stable-fast-3d.github.io.

Who it suits

The tool is aimed at developers and technical specialists who need fast mesh reconstruction with UV unwrapping from photos. It will suit those who work in a Python environment and have a GPU with CUDA or MPS support and sufficient VRAM. It will not suit users without command-line experience or those whose hardware does not meet the memory requirements. Also keep in mind that the code is distributed under the Stability AI Community License: commercial use is free only if annual revenue is under 1 million US dollars. Stable Fast 3D occupies its own niche among 3D generators: it does not require the cloud, but it places noticeable demands on local hardware. The last code change dates to January 2025, so the project is in an active phase of development.
Stable Fast 3D — generation pipeline From image to 3D GLB model Image One or more object views Preprocessing Normalization and alignment Reconstruction Mesh building and UV unwrapping Lighting Separation and texturing 3D model Export format GLB Ready asset Requirements VRAM: 6 GB System RAM: 32 GB Platform: CUDA Model weights: 3.7 GB
How the Stable Fast 3D pipeline works. The diagram is drawn from the tool’s fact sheet.

Fact sheet

Repository · Model on HuggingFace · Developer’s site

Taskimage-to-3d source
Code licenseStability AI Community License: commercial use free up to $1M in revenue source
Weights licenseStability AI Community License: free up to $1M in revenue source
PlatformCUDA, MPS per the author’s description source
VRAM6 GB per the author’s description source
System RAM32 GB per the author’s description source
Largest weights file3.7 GB source
All weights files3.7 GB source
Export formatsglb per the author’s description source
LanguagePython source
Last code change2025-01-22 source
Repository created2024-07-17 source
Changes often — as of 2026-09-14
GitHub stars1799 source
Forks227 source
Downloads per month12785 source
Model updated2025-04-08 source

Values are collected automatically from official sources and were checked on 2026-09-14. Each one links to its source, and values taken from the developer’s pages also carry a verbatim quote — hover over the note. Pricing and versions are shown as of the check date and change most often; verify on the vendor’s site before buying.

See also