AI Generation

Generate images from a model view or a Lite render to explore different looks. AI generation runs in the cloud. Connect to the internet and sign in to your D5 account, then choose AI Mode in the upper-left corner of Lite.

Getting a Base Image

• Use a view from your modeling software. Lite captures the current view when it opens. To use another view, switch to it in your modeling software, then click Capture from SketchUp, Capture from Rhino, or Capture from Revit in the lower-left corner of Lite. The button shows the name of your modeling software.

• Use a Lite render. Switch from Render Mode to AI Mode to use the current render as your starting image.

Controlling the Result

• Style Presets: built-in categories for architecture, landscape, and interior, each with finer project types and styles.

• Structure Weight: controls how closely the result follows the structure of the original. Lean toward Creative for more freedom (good for early ideation), or toward Accurate to stay closer to the original's lines and forms.

• Prompt: text describing what you want, up to 300 characters.

• Reference Image: add one reference image (up to 50 MB) to guide the style, color, or mood.

• Generation Settings: Image Quality (Standard / High), Number of Images (up to 4 at once), and an optional Fixed Seed.

Generate, Regenerate, and Iterate

• Generate: create the chosen number of images from the current starting image and settings.

• Regenerate: use when you want a different result or more variation; the input image stays the same and parameters can change.

• Iterate: use a generated image as the starting point for another round of changes.

Review and Save Results

1. When generation finishes, view the new results in AI mode.
2. Open History and click a record to review it. Lite displays the image and restores the base image and generation settings used for that result, so you can continue adjusting them.
3. Click the Save icon on the record's thumbnail to save the result image locally.

If generation is queued or fails, see AI generation troubleshooting.