neuralCAD-Edit - Model Quality: leaderboard
Metric: Expert model-quality rating (median of five CAD experts per edit on a 1-7 rubric, normalised to 0-1, times 100), on the 192 neuralCAD-Edit requests (expert CAD designers' video, speech, drawing and text edit requests on Autodesk Fusion models), the model editing the input STEP model by writing and re-running CadQuery scripts in the authors' harness with visual inspection; higher is better. Source: arxiv.org. Saturation forecast: Around September 2028. 3 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.2 | 39 |
| 2 | Gemini 3 Pro | 16 |
| 3 | Claude Sonnet 4.5 | 10 |
Interactive version: theaggregate.ai/benchmark?slug=neuralcad-edit-model-quality · How It Works · Data refreshed daily, snapshot 2026-10-07.