"Identity consistency test split frame layout" image Prompt + Comparison

multi-panel keyframe grid layout of this woman @img1 (4 panels total). image 1: her feet with black high heel shoes on from above, image 2: her skirt from waist till calves; image 3: her upper body waist up, image 4: equals @img1...

Log in to see full prompt.

This prompt tests:

Are exactly four panels visible in a clear grid layout?
Does each panel depict the same woman with consistent facial identity and features?
Is panel 1 a top-down view clearly showing her feet in black high heels?
Are the shoes in panel 1 correctly formed (no warped heels, straps, or extra toes)?
Does panel 2 show only the skirt region from waist to calves with correct proportions?
Is the skirt material consistent in color, texture, and lighting with other panels?
Does panel 3 show the upper body from waist up with no anatomical distortions?
Are hands and fingers in panel 3 correctly formed (no extra/missing fingers or warping)?
Does panel 4 accurately match the original reference image (@img1) without noticeable changes?
Is lighting direction and color consistent across all four panels?
Are body proportions coherent across panels (no scale or shape inconsistencies)?
Are panel boundaries clean with no visual bleeding or overlap between sections?

Check out the results from Freepik (Seedream 5.0 Lite) vs Freepik (FLUX.2 [pro]) vs Freepik (Nano Banana 2) vs ChatGPT's Image Generator by OpenAI (GPT Image 2.0) vs GROK (Grok Imagine v0.9 Image) for similar or identical prompts side-by-side.