Ad

text-to-cad: Generates CAD models with coding agents

text-to-cad enables generating 3D models using coding agents like Codex and Claude Code. It supports exporting to various CAD formats and offers a workflow for iterative design.
Screenshot of earthtojake/text-to-cad homepage

text-to-cad is an open-source harness designed to streamline the creation of 3D CAD models by leveraging the power of coding agents. The primary objective is to bridge the gap between natural language descriptions and precise CAD geometry. It addresses the challenge of manually creating complex models, making it faster and more accessible to generate CAD designs. The harness utilizes a combination of language models and CAD libraries such as OpenCascade to translate text prompts into manufacturable 3D models.

This project stands out by providing a comprehensive workflow for iterative CAD design. It includes features like reference copying for precise edits and visual review tools for quick feedback. The bundled skills for CAD, URDF, and robot motion offer a readily accessible set of tools, simplifying the setup process. The integration with CAD Explorer facilitates immediate visualization and inspection of generated geometry.

  • Model Export: Supports exporting to STEP, STL, 3MF, DXF, GLB, and URDF formats for wide compatibility.
  • Coding Agent Integration: Compatible with Codex and Claude Code for generating CAD models from text descriptions.
  • CAD Explorer: Provides an integrated CAD viewer for inspecting and reviewing generated models.
  • Reference Handling: Allows copying @[cad](/caddyserver/caddy)[...] references for precise, iterative edits.
  • Local Execution: Enables local execution without requiring a backend server.
  • Bundled Skills: Includes pre-built skills for CAD, URDF, and robot motion tasks.
  • Benchmark Suite: Contains a set of automated benchmarks for evaluating model generation capabilities.

The project is actively developed and maintained, with recent commits and ongoing issue resolution. The presence of a dedicated README, example benchmarks, and bundled skills indicates a growing community and a commitment to usability. While still evolving, the harness offers a solid foundation for AI-assisted CAD design and shows promising long-term potential.

text-to-cad benefits engineers, designers, and researchers seeking to accelerate the CAD design process. It is valuable for quickly prototyping 3D models, automating repetitive design tasks, and facilitating collaborative design workflows. By enabling text-based model creation, it reduces the reliance on manual CAD modeling, improving efficiency and reducing development time.

Summarize:
Share:
Stars
7,918
Forks
913
Issues
17
Created
3 months ago
Commit
14 days ago
License
MIT
Archived
No
Updated 13 days ago

Similar Repositories