Golden Hour Stroll
Vote
Image Editing
Photorealism
13 models were given the same image and edit instruction, and the community voted blind on which outputs looked best. How it works
#1 — Wan 2.6
Edit instruction
“Add dynamic motion to this photo: make hair blow in the wind, add leaves flying, energetic and lively feel.”
Challenge Rankings
| # | Model | Elo |
|---|---|---|
| 1 |
Wan 2.6
|
1269 |
| 2 |
Seedream 4.5
|
1267 |
| 3 |
GPT Image 1.5
|
1222 |
| 4 |
Nano Banana Pro
|
1210 |
| 5 |
Nano Banana
|
1201 |
| 6 |
Grok Imagine Image Pro
|
1191 |
| 7 |
Reve Image 1.0
|
1189 |
| 8 |
FLUX.2 [max]
|
1186 |
| 9 |
FLUX.2 [pro]
|
1181 |
| 10 |
Seedream 5.0 Lite
|
1178 |
| 11 |
Qwen Image Edit 2511
|
1177 |
| 12 |
Grok Imagine Image
|
1155 |
| 13 |
Seedream 4.0
|
1147 |
Wan 2.6 (1269 Elo) and Seedream 4.5 (1267 Elo) lead the challenge with win rates above 80%, establishing a 47-point Elo gap over the third-place GPT Image 1.5. Despite its premium price tier, Nano Banana Pro (1210 Elo) underperforms both GPT Image 1.5 and its significantly faster sibling, Nano Banana, which maintains a 62.5% win rate at a faster speed.
Elo vs Cost
Elo vs Speed
3 models waiting for enough speed data
Competitors
13 models, ranked by Elo
1
2
3
Highlighted Battles
The most competitive head-to-head matchups, selected by closeness and vote count.