Neutral Expression to Genuine Smile
Vote14 models were given the same image and edit instruction, and the community voted blind on which outputs looked best. How it works
#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
| # | Model | Elo |
|---|---|---|
| 1 |
Nano Banana
|
1278 |
| 2 |
Nano Banana Pro
|
1274 |
| 3 |
GPT Image 1.5
|
1270 |
| 4 |
Seedream 4.5
|
1245 |
| 5 |
Grok Imagine Image Pro
|
1240 |
| 6 |
Seedream 5.0 Lite
|
1234 |
| 7 |
Grok Imagine Image
|
1232 |
| 8 |
Qwen Image Edit 2511
|
1228 |
| 9 |
Nano Banana 2
|
1221 |
| 10 |
FLUX.2 [pro]
|
1192 |
| 11 |
FLUX.2 [max]
|
1185 |
| 12 |
Wan 2.6
|
1185 |
| 13 |
Seedream 4.0
|
1184 |
| 14 |
Reve Image 1.0
|
1107 |
Google's Nano Banana leads the challenge with 1278 Elo and a 73.6% win rate, outperforming its more expensive Pro counterpart despite a 72% faster generation time. GPT Image 1.5 closely trails the leaders at 1270 Elo, offering the highest value as the only budget-tier model in the top three while maintaining a 71.0% win rate.
Elo vs Cost
Elo vs Speed
3 models waiting for enough speed data
Competitors
14 models, ranked by Elo
1
2
3
Highlighted Battles
The most competitive head-to-head matchups, selected by closeness and vote count.