Back to browse
GitHub Repository

Autonomous multi-agent parametric design + CFD optimization using free local LLMs (Ollama), CadQuery, and OpenFOAM. $0 API cost.

0 starsPython

Autonomous CAD design and OpenFOAM optimization loop using local LLMs

by ostenjap·Jun 27, 2026·1 point·0 comments

AI Analysis

●●●BangerWizardryBold Bet

Physics referee validates AI designs—OpenFOAM CFD provides ground truth, not LLM self-evaluation.

Strengths
  • Multi-agent architecture with junior brainstormers, lead director, and physics referee actually works
  • Local Ollama LLMs mean zero API costs for the entire design optimization loop
  • CadQuery parametric CAD plus OpenFOAM CFD creates verifiable physical output
Weaknesses
  • Early stage demo—README shows research project, not polished production tool
  • Quadcopter propeller design is narrow domain despite impressive technical execution
Category
Target Audience

AI researchers, mechanical engineers, drone designers, agent system builders

Similar To

nTopology · Autodesk Fusion · Ansys

Post Description

LLM designs Quadcopter propeller and verifies with physics. It uses the Auto research Loop to find the best design.

Similar Projects