| scope | COMBO | auto | Auto follows the Paint tab: its Painted mode hands out the region, Whole frame hands out the picture, so the switch you set while painting is the one that runs. Whole frame hands out the full picture and a full-size mask, which is what most hosted models expect, and forces that even while painting a region. Painted region hands out just the painted area plus context, the way RedNode Paint Render works, which puts more pixels on the part being redone. |
| context | FLOAT | 0.250–2 | Painted region only: how much extra room around the paint to include, as a fraction of its size. Some context is what lets a renderer match the lighting instead of leaving a seam. |
| region_size | INT | 00–4096 | Painted region only: the PIXEL BUDGET the region is handed out at, as the side of a square. 1024 means about a megapixel however the region is shaped, so a long thin mask gets the same detail as a compact one instead of being rationed by its longest side. 0 follows the Paint tab's size dial, which is what the whole-frame path already does; anything else forces that budget. |
| region_shape | COMBO | auto | Painted region only: the shape the region is grown to. Auto picks whichever of 1:1, 4:3, 3:4, 16:9 and 9:16 the painted area is already closest to, which keeps it inside what the model was trained on. The box only ever GROWS, adding context around the paint, so nothing you painted is ever cropped out. |
| imageopt | IMAGE | | the picture to work from. Leave it unwired and it follows the Paint tab, the same as RedNode Paint Render does. |
| main_promptopt | STRING | | your main prompt. When the Paint tab's prompt box is empty, the prompt output hands this out instead, which is the same promise the box makes for the internal renderer. |
| main_negativeopt | STRING | | your main negative. When the Paint tab's negative box is empty, the negative output hands this out instead. |
