The Swap tab changes one thing in a photo and keeps everything else: the product, the outfit, the people, or the composition. Every swap starts from a Scene Photo, and the input mode decides what gets replaced.
How a swap works
Choose the input mode
Pick it under Input Mode in the settings panel. The model list follows the input mode.
Load the scene
Drop the photo to change on Scene Photo. It takes an image.
Load what goes in
The second slot is named after the input mode: see the table below.
Generate
The prompt is optional. Create writes the swap instruction, and anything you type is added to it as extra direction.
| Input mode | Second slot | What changes |
|---|---|---|
| Product | Product Asset | The product, at the same pose, angle, orientation and placement. |
| Product In Hand | Product Asset | The held product. The hand, fingers and grip adapt to it. |
| Outfit | Outfit Reference | The clothes only. The person stays the same. |
| Character | Character 1 and more | The people, keeping pose, lighting and framing. |
| Composition | Composition Reference | The framing, crop, zoom and point of view, taken from any reference image. |
The Outfit Reference, Composition Reference and Character slots also take a snippet. Its text then describes the outfit, the framing or the person in place of an image.
Character
Character replaces the people in the scene with the characters you load. The input panel offers up to five Character slots, fewer when the model accepts fewer images.
- With several people in the scene, the characters replace them from left to right: Character 1 takes the leftmost person.
- In the prompt, name them
@character1,@character2and so on. - Removing a character on the Input card moves the next ones up, and their mentions in the prompt follow.
Coherent Lighting
Coherent Lighting asks the model to match what it inserts (the product, the outfit or the characters) to the light of the scene: shadows, reflections and color. It has no effect in Composition.
Models
Each input mode offers its own models. The default model is ChatGPT Image 2.5 Flare.
Product and Product In Hand
OpenAI- ChatGPT Image 2.5 FlareDefault
- ChatGPT Image 2.5 Sunburst
- ChatGPT Image 2.0
GeminiGoogle- Nano Banana 2.1
- Nano Banana 2.0
- Nano Banana 2.0 Lite
- Nano Banana Pro
SeedreamBytePlus- Seedream 5.0 Pro
- Seedream 5.0 Flash
FluxBlack Forest Labs- Flux 3.0
Meta- Muse Image
GrokxAI- Grok Imagine 2.0
Ideogram- Ideogram 4.5
Outfit and Character
OpenAI- ChatGPT Image 2.5 FlareDefault
- ChatGPT Image 2.5 Sunburst
- ChatGPT Image 2.0
GeminiGoogle- Nano Banana 2.1
- Nano Banana 2.0
- Nano Banana 2.0 Lite
- Nano Banana Pro
SeedreamBytePlus- Seedream 5.0 Pro
- Seedream 5.0 Flash
FluxBlack Forest Labs- Flux 3.0
- Flux 2.0 Pro
- Flux 2.0 Max
- Flux Klein 2.0 9B
WanAlibaba- Wan Image 2.7
- Wan Image 2.7 Pro
- Wan Image 2.6
QwenAlibaba- Qwen 3.0 Pro
- Qwen 3.0
- Qwen 2.1 Pro
Meta- Muse Image
GrokxAI- Grok Imagine 2.0
Ideogram- Ideogram 4.5
Composition
OpenAI- ChatGPT Image 2.5 FlareDefault
- ChatGPT Image 2.5 Sunburst
- ChatGPT Image 2.0
GeminiGoogle- Nano Banana 2.1
- Nano Banana 2.0
- Nano Banana 2.0 Lite
- Nano Banana Pro
SeedreamBytePlus- Seedream 5.0 Pro
- Seedream 5.0 Flash
KlingKlingAI- Kling 3.0 Omni
WanAlibaba- Wan Image 2.7
- Wan Image 2.7 Pro
- Wan Image 2.6
FluxBlack Forest Labs- Flux 3.0
- Flux 2.0 Pro
- Flux 2.0 Max
Meta- Muse Image
GrokxAI- Grok Imagine 2.0