rayanht/msplat

Metal-accelerated 3D Gaussian Splatting for Apple Silicon

43
/ 100
Emerging

This tool helps 3D graphics professionals and researchers create highly realistic 3D scene representations from a collection of photos or videos. It takes standard camera data (like COLMAP, Nerfstudio, or Polycam datasets) and quickly generates a 3D Gaussian Splatting model. The output is a highly detailed, rendered 3D scene that can be viewed interactively or exported for use in other applications.

Available on PyPI.

Use this if you need to rapidly train high-quality 3D Gaussian Splatting models and render them at very high frame rates on Apple Silicon devices.

Not ideal if you are working on non-Apple hardware (like Windows or Linux machines with NVIDIA GPUs) or if you require a different 3D reconstruction method.

3D-reconstruction computer-graphics virtual-reality spatial-computing digital-twin
Maintenance 10 / 25
Adoption 5 / 25
Maturity 20 / 25
Community 8 / 25

How are scores calculated?

Stars

9

Forks

1

Language

Metal

License

Apache-2.0

Last pushed

Mar 10, 2026

Commits (30d)

0

Dependencies

1

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/rayanht/msplat"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.