Candid Street Photography
Vote
Text-to-Image
Photorealism
22 models were given the same prompt, and the community voted blind on which outputs looked best. How it works
#1 — Nano Banana Pro
Prompt
“A candid street photo of an elderly Japanese man repairing a red bicycle in light rain, reflections on wet pavement, shallow depth of field, 50mm lens, natural skin texture, imperfect framing, motion blur from passing cars, cinematic but realistic, no stylization.”
Challenge Rankings
| # | Model | Elo |
|---|---|---|
| 1 |
Nano Banana Pro
Google
|
1218 |
| 2 |
Seedream 4.5
ByteDance
|
1216 |
| 3 |
Grok Imagine Image Pro
xAI
|
1212 |
| 4 |
FLUX.2 [max]
Black Forest Labs
|
1201 |
| 5 |
Seedream 5.0 Lite
ByteDance
|
1178 |
| 6 |
GPT Image 1.5
OpenAI
|
1174 |
| 7 |
Qwen Image 2512
Alibaba
|
1173 |
| 8 |
GPT Image 1 Mini
OpenAI
|
1171 |
| 9 | ImagineArt 1.5 (Preview) Vyro AI | 1169 |
| 10 |
Imagen 4.0 Fast Generate 001
Google
|
1168 |
| 11 |
Seedream 4.0
ByteDance
|
1164 |
| 12 |
FLUX.2 [pro]
Black Forest Labs
|
1159 |
| 13 |
Grok Imagine Image
xAI
|
1156 |
| 14 |
Nano Banana
Google
|
1148 |
| 15 |
Nano Banana 2
Google
|
1147 |
| 16 |
Z-Image Turbo
Alibaba
|
1143 |
| 17 | Stable Diffusion 3.5 Large Stability AI | 1138 |
| 18 |
FLUX.2 [dev]
Black Forest Labs
|
1126 |
| 19 | FLUX.2 [dev] Turbo fal | 1124 |
| 20 |
FLUX.2 [flex]
Black Forest Labs
|
1119 |
| 21 | FLUX.2 [dev] Flash fal | 1106 |
| 22 |
Imagen 4.0 Ultra Generate 001
Google
|
1078 |
Elo vs Cost
Elo vs Speed
9 without speed data omitted.
Highlighted Battles
The most competitive head-to-head matchups, selected by closeness and vote count.
Competitors
22 models, ranked by Elo
1
2
3