When I provide a blueprint or measurements, the AI rendering should follow my actual layout — not invent its own. Right now, objects get placed on the wrong walls, elements are misidentified (for example, a cabinet gets called a doorway), and the layout is ignored even when I upload a blueprint and provide measurements.
I have to make multiple attempts just to get close to what I actually want. It would be a lot more useful if the AI read my spatial context — the blueprint I uploaded, the dimensions I gave — and used that as the foundation for the rendering.
I compared the same inputs in Grok and got significantly better results. Handoff AI needs to catch up here.
Please authenticate to join the conversation.
Submitted
💡 Feature Requests
8 days ago

Richard Lawson
Get notified by email when there are changes.
Submitted
💡 Feature Requests
8 days ago

Richard Lawson
Get notified by email when there are changes.