Generates 3D models from text descriptions or images using AI image generation, multi-view reconstruction, and OpenSCAD.
OpenSCAD enables users to generate 3D models from text or images, leveraging AI-powered image generation through Google Gemini or Venice.ai APIs. It employs multi-view image generation, allowing for 3D reconstruction and an image approval workflow. The server integrates CUDA Multi-View Stereo for 3D reconstruction and generates parametric 3D models with OpenSCAD. Remote processing capabilities offload intensive tasks to LAN servers. The system supports parametric export in various formats and optional 3D printer discovery, all managed through a modular Python MCP SDK architecture.
Key Features
01OpenSCAD Integration for Parametric 3D Models
02Multi-View Image Generation and 3D Reconstruction
033 GitHub stars
04AI Image Generation using Google Gemini or Venice.ai
05Automatic LAN server discovery for CUDA-MVS
06Remote Processing for CUDA Multi-View Stereo
Use Cases
01Generating 3D models from text prompts
02Creating parametric designs for 3D printing
03Offloading computationally intensive 3D reconstruction tasks
