Modly Image to 3D Skill Skill by [ara.so](https://ara.so) — Daily 2026 Skills collection. Modly is a local, open source desktop application (Windows/Linux) that converts photos into 3D mesh models using AI models running entirely on your GPU — no cloud, no API keys required. Architecture Overview The app runs as an Electron shell over a local Python FastAPI server. Extensions are GitHub repos with a manifest.json + generator.py that plug into the extension system. Installation Quick start (no build required) Development setup Production build Key npm Scripts Extension System Extensions are GitHub repositories containing: manifest.json — metadata and model variants generator.py — generation logic implementing the Modly extension interface manifest.json structure generator.py interface Installing an extension (UI flow) 1. Open Modly → go to Models page 2. Click Install from GitHub 3. Paste the HTTPS URL, e.g. https://github.com/lightningpixel/modly hunyuan3d mini extension 4. After install, click Download on the desired model variant 5. Select the installed model and upload an image to generate Official Extensions Extension Model [modly hunyuan3d mini extension](https://github.com/lightningpixel/modly hunyuan3d mini extension) Hunyuan3D 2 Mini Python Backend API (FastAPI) The backend runs locally. Key endpoints used by the Electron frontend: Calling the backend from Electron (IPC pattern) Writing a Custom Extension Minimal extension repository structure Example: wrapping a HuggingFace diffusion model Configuration & Environment Modly runs fully locally — no environment variables or API keys needed. GPU/CUDA is auto detected by PyTorch in extensions. Relevant configuration lives in: If you need to configure the backend port or extension directory, check the Electron main process config (typically src/main/index.ts ) for constants like API PORT or EXTENSIONS DIR . Common Patterns Check if CUDA is available in an extension Progress reporting from generator.py Adding a new page in the renderer (React) Troubleshooting Problem Fix npm run dev — Python backend not starting Ensure venv is set up: cd api && python m venv .venv && pip install r requirements.txt CUDA out of memory Use a smaller model variant or close other GPU processes Extension install fails Verify the GitHub URL is HTTPS and the repo contains manifest.json at root Generation hangs Check that your GPU drivers and CUDA toolkit match the PyTorch version in requirements.txt App won't launch on Linux Make launcher.sh executable: chmod +x launcher.sh Model download stalls Check disk space; large models (4–10 GB) need adequate free space torch not found in extension Ensure PyTorch is in api/requirements.txt , not just the extension's own deps Verifying GPU is detected Resources Homepage : https://modly3d.app Releases : https://github.com/lightningpixel/modly/releases/latest Official extension : https://github.com/lightningpixel/modly hunyuan3d mini extension Discord : https://discord.gg/FjzjRgweVk License : MIT (attribution required — credit Modly + Lightning Pixel in forks)

modly-image-to-3d