neuralCAD-Edit: An Expert Benchmark for Multimodal-Instructed 3D CAD Model Editing
For researchers developing 3D CAD editing methods and foundation models, this benchmark provides a realistic and challenging evaluation standard.
The paper introduces neuralCAD-Edit, the first benchmark for editing 3D CAD models using multimodal instructions from expert designers. It reveals a large performance gap, with the best foundation model scoring 53% lower than human experts in acceptance trials.
We introduce neuralCAD-Edit, the first benchmark for editing 3D CAD models collected from expert CAD engineers. Instead of text conditioning as in prior works, we collect realistic CAD editing requests by capturing videos of professional designers, interacting directly with CAD models in CAD software, while talking, pointing and drawing. We recruited ten consenting designers to contribute to this contained study. We benchmark leading foundation models against human CAD experts carrying out edits, and find a large performance gap in both automatic metrics and human evaluations. Even the best foundation model (GPT 5.2) scores 53% lower (absolute) than CAD experts in human acceptance trials, demonstrating the challenge of neuralCAD-Edit. We hope neuralCAD-Edit will provide a solid foundation against which 3D CAD editing approaches and foundation models can be developed. Code/data: https://autodeskailab.github.io/neuralCAD-Edit