# CADWorld: A New Benchmark for Long-Horizon Computer-Aided Design

Digest AI · Generative AI & Models · published 2026-09-16T04:00:00Z

Canonical: https://digestai.news/story/cadworld-a-new-benchmark-for-long-horizon-computer-aided-design

## Summary

Computer-use agents are increasingly evaluated in realistic desktop environments, but benchmarks often lack coverage of professional engineering workflows. This article introduces CADWorld, a new benchmark specifically designed for long-horizon computer use in FreeCAD. It contains over 200 tasks across eleven mechanical-CAD workflow categories, including sketching, part modeling, assembly, CAM (Computer-Aided Manufacturing), FEM (Finite Element Method), measurement, mesh processing, and technical drawing. Agents operate through screenshots and GUI actions, with success determined by task-specific checks on saved FreeCAD artifacts and auxiliary outputs such as geometric properties, parametric structure, constraints, manufacturing state, and simulation results. The strongest agent achieves only 17.5% success rate compared to the 87.0% expert reference pass. This benchmark highlights a gap between general GUI competence and reliable execution of persistent engineering workflows.

## Key points

- CADWorld introduces a new benchmark for long-horizon computer use in FreeCAD
- Contains over 200 tasks across eleven mechanical-CAD workflow categories
- Success determined by checks on saved FreeCAD artifacts and auxiliary outputs

## Why it matters

This benchmark exposes a significant gap between general GUI competence and reliable execution of persistent engineering workflows, which is crucial for the development and evaluation of computer-aided design agents.

## Sources

1. [CADWorld: Computer-Use Benchmark for Long-Horizon Computer-Aided Design](https://arxiv.org/abs/2609.16251) (arXiv cs.AI, 2026-09-16, primary source)

## Cite

Digest AI, "CADWorld: A New Benchmark for Long-Horizon Computer-Aided Design", 16 September 2026, https://digestai.news/story/cadworld-a-new-benchmark-for-long-horizon-computer-aided-design

---

Written by Digest AI's editorial model from the linked sources; the sources are the record. Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse
JSON: https://digestai.news/story/cadworld-a-new-benchmark-for-long-horizon-computer-aided-design.json