generationstringrequiredThe id of the set to edit: the random segment of its download URL, as list_recent_generations returns it. One batch of a set per call.
A set you already generated, delivered again at a different size, margin, sizing, file type, quality, or, for a transparent set, on any background color: edit_image_set re-cuts it from its full-resolution source. Free to call.
edit_image_setGeneration decides what the pictures are. Everything after that, the size they arrive at, the margin around each subject, the file type and its quality, and for a transparent set the color under it, is a delivery decision, and this tool lets you change it later without spending a credit or getting a different set. The set's download is replaced in place: the same URL now serves the new delivery. A lever you leave out keeps its current value, so one call can change one thing, and any earlier state, the original included, is one call away.
# Deliver an existing set as a 512 x 512 WebP instead
edit_image_set(
generation: "8a617b9dd9cf51bed87878cc",
width: 512,
height: 512,
format: "webp",
quality: 85
)
# Put a transparent set on a dark surface, then back to transparent
edit_image_set(generation: "8a617b9dd9cf51bed87878cc", background: "#111318")
edit_image_set(generation: "8a617b9dd9cf51bed87878cc", background: "transparent")Or, in plain language to your AI:
Use Logospell to deliver my last image set again as 512 by 512 WebP at quality 85, then once more composed over #111318.generationstringrequiredThe id of the set to edit: the random segment of its download URL, as list_recent_generations returns it. One batch of a set per call.
widthintegeroptionalNew delivered width, 256 to 512, or 0 to stop fixing the width. Combines with height exactly as on the set tools: both fixed is an exact box, one fixed lets the other hug each subject, neither is native size.
heightintegeroptionalNew delivered height, 256 to 512, or 0 to stop fixing the height.
canvasstringoptionalWith neither width nor height fixed, what the images land on: uniform (the default) puts the whole set on one canvas, the size it was generated at, every image alike; subject wraps each image around its own subject, so the images differ in size the way the subjects do. Says nothing when a size is fixed.
minimumMarginintegeroptionalNew minimum margin in pixels around each subject, with the set tools' bounds.
formatstringoptionalpng, webp, or jpg. jpg has no alpha: on a transparent set it composes the images over white unless a background color is given in the same call.
qualityintegeroptional1 to 100 for webp and jpg; ignored for png.
backgroundstringoptionalTransparent sets only: a #RRGGBB color composes every image over it and delivers an opaque set; "transparent" restores the alpha. Rejected on a solid-background set, whose color is in its pixels.
sizingstringoptionalrelative keeps the sizes the model gave the subjects in relation to one another (one scale for the set); fill scales each subject on its own to fill the frame less the margin, so every image reads at the same visual weight, at the cost of relative size and of enlarging subjects smaller than the frame (the result reports by how much).
Style, subjects and references are generation, not delivery: a different picture is a new set. Editing never extends a set's download window, and a set generated before this tool existed has no source to cut from. The same controls are on each set on the generations page.