# MrNeRF/LichtFeld-Studio

Train, inspect, edit, automate, and export 3D Gaussian Splatting scenes from a single native application.

Repository: https://github.com/MrNeRF/LichtFeld-Studio
Canonical: https://ross.abutalabs.com/products/lichtfeld-studio
Homepage: https://lichtfeld.io
Language: C++
License: GPL-3.0
License Family: copyleft
Topics: cuda, gaussian-splatting, optimization, computer-graphics, computer-vision
Last push: 2026-08-26T21:49:19+00:00

## Health v2 (maintenance only)
Score: 92/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 90, longevity 80
- inputs: {"age_days": 1130, "days_push": 7, "days_rel": 70, "gap_med": 16.5, "n_releases_24m": 7}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3594, forks 432 (observed 2026-08-28T04:08:11.217605+00:00)

## What it is
LichtFeld Studio is a native open-source desktop application for 3D Gaussian Splatting that combines training, real-time inspection, splat editing, and export in one toolchain. It is built on a C++23/CUDA core and is extensible via Python plugins and MCP automation endpoints.

## Use cases
- train 3D gaussian splatting scenes from COLMAP datasets
- view and inspect 3DGS reconstructions in real time
- edit and clean up gaussian splat scenes with brush and lasso tools
- export gaussian splatting scenes to PLY, SOG, SPZ, or HTML viewer
- automate 3DGS training workflows with Python scripts or MCP
- resume 3DGS training from checkpoints
- render photorealistic 3D scenes from photos or video locally on GPU

## When to choose
- you want an all-in-one native GUI for the full 3DGS pipeline instead of separate training scripts and viewers
- you need to interactively edit, select, and transform gaussian splats with undo/redo
- you want to extend or automate 3DGS workflows via Python plugins or MCP
- you need fast CUDA-based training with checkpoint resume on Windows or Linux

## When to avoid
- you don't have a CUDA-capable NVIDIA GPU
- you need a headless, script-only training pipeline without a GUI
- you need non-GPLv3 licensing for proprietary redistribution
- you work on macOS, which is not supported

## Facets
- artifact type: application
- maturity: active
- function: machine-learning, computer-vision, graphics, gui, plugin-system, mcp, image-processing, gpu-computing
- domain: computer-vision, graphics, machine-learning, desktop-applications, developer-tools
- platform: windows, cpp, python
- tags: gaussian-splatting, 3dgs, 3d-reconstruction, photogrammetry, scene-editing, neural-rendering, colmap, cuda, mcp-automation, linux, desktop

## Member repositories
- MrNeRF/LichtFeld-Studio (main) score 92

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:11.217605+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 2026-08-29T18:33:48.881695+00:00, confidence not recorded.
  - readme: https://github.com/MrNeRF/LichtFeld-Studio (fetched 2026-08-28T04:08:11.217605+00:00, sha 568015489b11)
  - homepage: https://lichtfeld.io (fetched 2026-08-29T09:27:07.047442+00:00, sha f97fd4ce7b0a)
  - site_page: https://lichtfeld.io/docs (fetched 2026-08-29T09:27:07.050099+00:00, sha d217c848d4e7)
  - site_page: https://lichtfeld.io/blog/lichtfeld-studio-session-1 (fetched 2026-08-29T09:27:07.052016+00:00, sha fa8b93b85775)
- Data as of 2026-08-30T08:39:29.467469+00:00.