Neutral Expression to Genuine Smile

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

Blind Vote This Challenge

Winning image by Nano Banana for Neutral Expression to Genuine Smile

#1 — Nano Banana

Edit instruction

{
  "action": "image_edit",
  "reference": "uploaded neutral portrait",
  "change": "Warm genuine Duchenne smile: lips curved up, slight natural teeth, soft eye crinkles, subtle cheek raise",
  "details": "Realistic smiling skin (dimples if present, soft cheek shadows), slightly brighter eyes; keep exact eye shape/color/iris",
  "preserve_exact": "Face identity/structure, eyes/nose/lips/eyebrows, hair, skin texture/pores/freckles, makeup, clothing, head pose, background, lighting, shadows, framing",
  "no_changes": "No face shape change, no new features, no gaze shift, no hair/clothing/lighting/background edits",
  "style": "Ultra-photorealistic 8K portrait, sharp face focus, natural soft lighting, realistic skin glow"
}

Voters were asked to judge by Identity preserved Natural smile No geometry changes

Challenge Rankings

14 models

#	Model	Price	¢/img	Elo
1	Nano Banana Google	$$	3.9¢	1287
2	GPT Image 1.5 OpenAI	$	0.9¢	1285
3	Nano Banana Pro Google	$$$	6.7¢	1274
4	Seedream 4.5 ByteDance	$$	4¢	1245
5	Qwen Image Edit 2511 Alibaba	$$	3¢	1235
6	Grok Imagine Image xAI	$$	2¢	1232
7	Nano Banana 2 Google	$$	2.2¢	1221
8	Seedream 5.0 Lite ByteDance	$$	3.5¢	1218
9	Grok Imagine Image Pro xAI	$$$	7¢	1215
10	Wan 2.6 Alibaba	$$	3¢	1185
11	Seedream 4.0 ByteDance	$$	3¢	1172
12	FLUX.2 [max] Black Forest Labs	$$	3¢	1170
13	FLUX.2 [pro] Black Forest Labs	$$	1.5¢	1163
14	Reve Image 1.0 Reve AI	$$	4¢	1108

Google’s Nano Banana leads with a 75% win rate (1287 Elo), narrowly edging out OpenAI’s GPT Image 1.5 (1285 Elo) which delivers nearly identical performance at less than one-fourth the price. While Nano Banana Pro holds a top-three position (1274 Elo), the standard Nano Banana model remains more effective and over twice as fast (8s vs 20.9s) for this portrait editing task.