Here is a real use case for a free Apple tool most people are treating as a toy. Apple released SHARP, an open research model that turns one photo into a 3D Gaussian splat scene. Not a flat picture. A scene you can rotate. That matters on a real job, because once drywall goes up, the MEP inside the wall is gone from view. A photo helps. A rotatable 3D record helps more, if you capture it right and you do not pretend it is something it is not. This is a step-by-step tutorial, field first. https://github.com/apple-aiml-research/ml-sharp?utm_source=syndication&pubDate=20261104 What SHARP is: SHARP is Apple's free, open research model. It runs offline and locally, and it lives on GitHub. It turns a single photo into a 3D Gaussian splat scene, built from millions of tiny, soft, semi-transparent blobs of color that blend into a photorealistic image you can rotate. The output is metric, with real-world scale, and it saves as a standard .ply file that works with public splat viewers. Each scene is about 1.2 million Gaussians, in two 768 by 768 layers. It is the full-3D cousin of iOS 26 Spatial Scenes, which only shifts a subject against a background when you tilt the phone. What SHARP is not: It is not a measurement tool, and it is not a BIM replacement. It estimates depth and 3D structure from one view. What the camera cannot see, it has to guess. It does not prove what is behind a pipe, inside a wall cavity, or on the far side of a stud. Use it as a visual record, tied to a real QA/QC check. Never as proof of hidden work by itself. And one honest limit, straight from the article: SHARP runs on an Apple silicon Mac only. It does not run natively on an iPhone today. So the iPhone is your capture and viewing device. The Mac does the 3D processing. Your app ties the two together. The workflow, in plain terms: Capture the open wall, turn each key photo into a 3D splat, attach it to the room and wall, check it, sign it off, then keep it for Operations.