Outfit Transfer Challenge

Image Editing Photorealism

8 models were given the same image and edit instruction, and the community voted blind on which outputs looked best. How it works

This is extremely practical (think e-commerce, fashion, virtual dressing rooms). It’s visually very obvious when it fails, and success looks impressive. It tests a model’s understanding of clothing physics, body shape, and lighting.

Blind Vote This Challenge

#1 — Nano Banana Pro

Edit instruction

“Use Image 1 as the base person. Dress them in the exact elaborate outfit from Image 2 (including all layers, accessories, jewelry, and shoes). Carefully adapt the clothing to the body shape and pose in Image 1 while maintaining realistic fabric behavior, correct proportions, and perfect lighting/shadow matching. Keep the person’s exact face, hair, and background completely unchanged.”

Voters were asked to judge by Clothing Accuracy Fit & Fabric Realism Lighting & Shadow Consistency No Side Effects

Challenge Rankings

8 models

#	Model	Price	¢/img	Speed	Elo
1	Nano Banana Pro Google	$$$	6.7¢		1182
2	FLUX.2 [klein] 9B Black Forest Labs	$	0.6¢		1135
3	GPT Image 2 OpenAI	$$	3.9¢		1132
4	GPT Image 1 OpenAI	$$	1.1¢		1127
5	Wan 2.7 Alibaba	$$	3¢		1095
6	GPT Image 1 Mini OpenAI	$	0.5¢		1082
7	Qwen Image Edit Latest Alibaba	$$$	7.5¢	—	1062
8	Grok Imagine Image xAI	$$	2¢		996

Nano Banana Pro leads the leaderboard with a 100% win rate and an Elo of 1182, establishing a 47-point gap over the budget-tier FLUX.2 [klein] 9B. Despite its low price point and nearly 10x faster generation speed, FLUX.2 [klein] 9B (1135 Elo) narrowly outperforms expensive competitors like GPT Image 2 and Qwen Image Edit Latest in complex fashion transfer tasks.