Turn Any Instagram Carousel Template Into Your Own

What Ribbi did

Ribbi starts with a structured breakdown of the reference — the three-panel vertical narrative (desk work → close-up → lying flat), the sepia-red and teal-gray duotone, the hand-drawn linework and grain, the burned-out urban mood — then picks nano-banana-pro in style-transfer mode, feeding the full image analysis as the prompt with the original page as the style source. One pass produces a 1696x2528 recreation with the panel structure, palette, brushwork and mood all locked in, ready to post.

What Ribbi did

Ribbi first collapses the request into three variables — style, theme, count — then designs a three-panel storyline for each: "commute hell" with the subway crush and surging crowds, "late-night convenience store" with fluorescent light and street-bench noodles, and "weekend shut-in" sliding from a 1 p.m. alarm to phone-scrolling after dark. Each comes with panel angles and a warm-cold color scheme. After a quick confirmation, all three 1696x2528 pieces are generated in one batch, reading as a series while standing alone.

What Ribbi did

Ribbi first checks whether all three should share one color or each get its own, then regenerates the set in teal: the asymmetric three-panel split, sketchy linework and grain texture stay intact while the color system collapses to a single teal scale, and the emotional camera language — enclosure → close-up → release — carries through, so the three themes read as one visual family.

What Ribbi did

Ribbi draws the line between regenerating and cropping, confirms the target is the teal monochrome batch, then switches to an image editor's edit mode for a pure crop: all three go from 2:3 portrait to 2048x2048 squares with the compositional focus preserved and zero changes to the artwork — drop-in ready for the carousel.

What Ribbi did

Ribbi decomposes the reference dimension by dimension into a complete English prompt — composition (asymmetric three-panel split, square format), style (hand-drawn sketch, crosshatching, indie-comic aesthetic), color (teal monochrome, heavy shadows, film grain) — then runs Midjourney v7 with the dual-reference setup, story from one image and style from the other, producing four 1024x1024 variants to choose from. The same visual language reproduces cleanly on a different model.