CADReasoner: Iterative Program Editing for CAD Reverse Engineering

TL;DR

CADReasoner is an iterative AI model that refines CAD reverse engineering by using geometric discrepancies and multi-modal data, achieving state-of-the-art results on multiple benchmarks.

Contribution

It introduces an iterative refinement approach with a new scan-simulation protocol, improving reliability and detail in CAD reverse engineering.

Findings

Achieves state-of-the-art results on DeepCAD, Fusion 360, and MCB benchmarks.

Effectively fuses multi-view renders and point clouds for better shape reconstruction.

Outperforms previous single-pass and agent-based methods in CAD reverse engineering.

Abstract

Computer-Aided Design (CAD) powers modern engineering, yet producing high-quality parts still demands substantial expert effort. Many AI systems tackle CAD reverse engineering, but most are single-pass and miss fine geometric details. In contrast, human engineers compare the input shape with the reconstruction and iteratively modify the design based on remaining discrepancies. Agent-based methods mimic this loop with frozen VLMs, but weak 3D grounding of current foundation models limits reliability and efficiency. We introduce CADReasoner, a model trained to iteratively refine its prediction using geometric discrepancy between the input and the predicted shape. The model outputs a runnable CadQuery Python program whose rendered mesh is fed back at the next step. CADReasoner fuses multi-view renders and point clouds as complementary modalities. To bridge the realism gap, we propose a scan-simulation protocol applied during both training and evaluation. Across DeepCAD, Fusion 360, and MCB benchmarks, CADReasoner attains state-of-the-art results on clean and scan-sim tracks.