# CADEvolve: Creating Realistic CAD via Program Evolution

arXiv: https://arxiv.org/abs/2602.16317 · HTML: https://mlsanity.com/paper/2602.16317/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 18 Feb 2026.

## Built on by (most cited first, 7 of 7)

- [CADBench: A Multimodal Benchmark for AI-Assisted CAD Program Generation](https://mlsanity.com/paper/2605.10873/) (2026)
- [ArtiCAD: Articulated CAD Assembly Design via Multi-Agent Code Generation](https://mlsanity.com/paper/2604.10992/) (2026)
- [CADEngBench: It Looks Like CAD, but Does It Work? Evaluating Parametric Design, Assembly Reasoning, and Physics Simulation](https://mlsanity.com/paper/2608.09296/) (2026)
- [Pointer-CAD v2: Plan-Then-Construct CAD Generation with Dimension-Aware Parametric Precision](https://mlsanity.com/paper/2606.29301/) (2026)
- [CADENA: Stepwise CAD Reverse Engineering](https://mlsanity.com/paper/2608.00799/) (2026)
- [StepCAD: Mesh-to-CAD Code Generation via LLM Policy and Geometry-Guided Search](https://mlsanity.com/paper/2610.03799/) (2026)
- [CADFather: Autonomous CAD Reconstruction through Coordinated Tool Use](https://mlsanity.com/paper/2610.09127/) (2026)

## Abstract (arXiv)

Computer-Aided Design (CAD) delivers rapid, editable modeling for engineering and manufacturing. Recent AI progress now makes full automation feasible for various CAD tasks. However, progress is bottlenecked by data: public corpora mostly contain sketch-extrude sequences, lack complex operations, multi-operation composition and design intent, and thus hinder effective fine-tuning. Attempts to bypass this with frozen VLMs often yield simple or invalid programs due to limited 3D grounding in current foundation models. We present CADEvolve, an evolution-based pipeline and dataset that starts from simple primitives and, via VLM-guided edits and validations, incrementally grows CAD programs toward industrial-grade complexity. The result is 8k complex parts expressed as executable CadQuery parametric generators. After multi-stage post-processing and augmentation, we obtain a unified dataset of 1.3m scripts paired with rendered geometry and exercising the full CadQuery operation set. A VLM fine-tuned on CADEvolve achieves state-of-the-art results on the Image2CAD task across the DeepCAD, Fusion 360, and MCB benchmarks.