Style Spectrum: From the Same Realistic Illustration to Flat Gradient

Using a single realistic life-scene illustration as the sole reference image, this work leverages gpt-image-2's image input capability to reinterpret it into three styles—mid-tone, semi-flat, and fully flat—while preserving the main subject composition and narrative. The piece presents layered results along the same "realistic→flat" style spectrum, making it easy for developers to observe the controllable range of style intensity in image-to-image, as well as the differences in handling lines, color blocks, shadows, and detail density. Applicable scenarios include illustration style exploration, visual concept comparisons, and style gradient validation for design systems.

Style Illustration · style-transfer · Flat design

Model GPT Image 2

Reference inputs

Reference images

Try it yourself

Prompt

Try this model

A set of illustration series showing a style spectrum from realistic to flat. Starting from a warm, softly lit realistic indoor scene in which a woman reads by a window, reinterpret it across three stylization levels: (1) semi-realistic, with soft painterly gradients and retained detail; (2) semi-flat, with simplified shadows, bold silhouettes, and reduced texture; (3) fully flat vector style, using solid color blocks, clean geometric shapes, and minimal lines. Keep the subject, composition, and color palette consistent across all levels. 16:9 composition, clean edges, no text.

Copy the full prompt and adapt it for your own generation.