Upload reference images and apply natural-language edits while preserving lighting and composition. Supports multi-image inputs and high-res 4K outputs.
Approx. Price
$0.140 per image
Model Type
image-to-image
Preview Examples
6
Generation controls available for this model.
Images Per Run
Up to 10
Resolution Options
3
1k (Fast edits (~1024px class)), 2k (Balanced detail (~2048px class)), 4k (Maximum fidelity (~4096px class))
Tunable Settings
3
Aspect Ratio
Default
auto
Options (11)
Auto (match input), Square, 3:2, 2:3 +7 more
Set the target framing for edits.
Number of Images
Default
1
Resolution
Default
1k
Options (3)
1k (Fast edits (~1024px class)), 2k (Balanced detail (~2048px class)), 4k (Maximum fidelity (~4096px class))
Human preference benchmarks sourced from Artificial Analysis.
No Artificial Analysis benchmark data is available yet for this model.
Artificial Analysis API
Live action movie still, Japanese youth film style. A beautiful young woman in a collegiate outfit running through a campus gate, laughing. Dynamic motion, hair flying. Surrounded by falling pink cherry blossom petals. Soft sunlight, airy atmosphere, shallow depth of field, bokeh background of other students, Fujifilm color tone, realistic, sharp focus on eyes.
Translate the text in the image into Portuguese from brazil.
Make the hamburger made of glass.
Place the sofa from Figure 2 into Figure 1, and blend the light and shadows.
Turn the image into Minecraft style.
Turn this scene into nighttime.
