# lightningpixel/modly

Desktop app to generate 3D models from images or prompt using local AI — runs entirely on your GPU

Repository: https://github.com/lightningpixel/modly
Canonical: https://ross.abutalabs.com/products/modly
Homepage: https://modly3d.app
Language: TypeScript
License: NOASSERTION
License Family: other
Topics: 3d, ai-local, ai-tools, open-source, modly, desktop-app, self-hosted
Last push: 2026-08-26T17:14:24+00:00

## Health v2 (maintenance only)
Score: 80/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 93, longevity 12
- inputs: {"age_days": 169, "days_push": 7, "days_rel": 48, "gap_med": 6, "n_releases_24m": 12}
- flags: young, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 7275, forks 689 (observed 2026-08-28T04:09:57.938638+00:00)

## What it is
Modly is an open-source desktop application that generates textured 3D meshes from images or text prompts using open-source AI models (Hunyuan3D, TripoSG, Trellis) running entirely on your local GPU. It supports an extension system for adding new models, mesh smoothing/decimation, and exports to GLB, OBJ, STL, and PLY formats.

## Use cases
- turn a photo into a 3D model
- generate 3D meshes from text prompts locally
- create game assets for Unity or Unreal from images
- convert images to 3D models for 3D printing
- run image-to-3D AI models offline without the cloud
- prepare and export 3D models for Blender
- swap in custom image-to-3D models via extensions

## When to choose
- You want private, offline image-to-3D generation with no uploads or credits
- You have an NVIDIA GPU (or Apple Silicon) and want local AI 3D mesh generation
- You need 3D assets in standard formats for game engines, Blender, or 3D printing
- You want a hackable, extensible pipeline for open-source image-to-3D models

## When to avoid
- You have no compatible GPU and cannot run local inference
- You need cloud-scale throughput or batch generation on servers
- You need production-grade CAD or precise parametric modeling rather than AI-generated meshes
- You require a permissively-licensed dependency — the license is listed as NOASSERTION

## Facets
- artifact type: application
- maturity: active
- function: machine-learning, image-processing, graphics, gui, plugin-system
- domain: artificial-intelligence, graphics, desktop-applications, self-hosted
- platform: windows, python
- tags: image-to-3d, text-to-3d, 3d-mesh-generation, local-ai, hunyuan3d, triposg, trellis, electron, offline, gpu-inference, 3d-printing, blender, game-development, linux, macos, desktop, gpu, typescript

## Member repositories
- lightningpixel/modly (main) score 80

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:57.938638+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 2026-08-29T17:38:54.511150+00:00, confidence not recorded.
  - readme: https://github.com/lightningpixel/modly (fetched 2026-08-28T04:09:57.938638+00:00, sha 01178fe2d093)
  - homepage: https://modly3d.app (fetched 2026-08-29T08:34:29.511668+00:00, sha 98f3203e623b)
  - site_page: https://modly3d.app/docs (fetched 2026-08-29T08:34:29.520986+00:00, sha a5f4fc90e39d)
- Data as of 2026-08-30T08:39:29.467469+00:00.