What It Does

This tool generates new images based on uploaded references and a selected model.

It allows you to:

edit or transform existing images
combine multiple references into one result
apply LoRA-based styles
generate consistent variations across images

Result examples:

Interface Overview

Images

Upload one or more images to use as references.

Up to 100 images per run
Images can be used together or processed separately (Batch Mode)

Model

Select the model that will generate the result.

Available models:

General (4K, Minimal filters)
Highest quality model, best for final images
Nano Banana 2 (2K, ON filters)
Clean, brand-safe and photorealistic results (SFW-focused)
Qwen Image 2.0 / Pro (2K, Minimal filters)
Balanced models for structured and consistent editing
WAN 2.7 / Pro (Minimal filters)
More flexible models for varied and expressive results
Flux Klein (1–2MP, ON filters)
Fast model for quick edits and iteration
Flux Klein NSFW (1–2MP, Minimal filters)
Version with fewer content restrictions
Flux Klein + LoRA (1–2MP, Minimal filters)
Supports LoRA style templates

Prompt (optional)

Describe the result you want to generate.

The prompt helps control:

appearance
style
composition

Batch Mode

When enabled, each uploaded image is processed separately.

Each image becomes its own task
The same prompt is applied independently
Images do not influence each other

Use Batch Mode when:

you want consistent edits across multiple images
you don’t want images to be combined

LoRA Style Templates

LoRAs allow you to apply predefined styles and visual patterns.

Selecting a LoRA applies a specific style
You can still add your own prompt, but it is recommended to keep it short, as long or complex prompts may affect LoRA stability.

Additional Settings

Aspect Ratio — sets the width-to-height ratio of the output (e.g., 1:1, 3:4, 9:16)

Resize (W / H) — manually controls image resolution in pixels

Number of Images — defines how many variations to generate (up to 10)

Sequential Generation (available in General model) — generates a consistent series of images with the same characters, style, and environment

Generate

Starts the generation process.

Credit cost is shown before generation
Cost depends on model and number of images

Highlights

Highlights

Upload up to 100 input images per run
Supports multiple models for different use cases (General, Nano Banana 2, Qwen, WAN, Flux)
Works with minimal Safety Filters (General, WAN, Flux NSFW) — supports both SFW and NSFW content
Allows combining multiple references or processing them separately (Batch Mode)
Supports LoRA style templates for consistent visual results
Seamless integration with Upscale, Face Swap, Video, and Carousel tools

Model selection support

Use General for highest quality final images
Use Nano Banana 2 for clean, brand-safe SFW visuals
Use Qwen for structured and consistent editing
Use WAN for more flexible and expressive results
Use Flux Klein for fast iteration and quick edits
Use Flux Klein NSFW for fewer content restrictions
Use Flux Klein + LoRA for style-based generation

Pro Tips

Always add a clear face image to preserve identity
Include a full-body photo for better pose and proportion accuracy
Use Batch Mode when applying the same edit to multiple images independently
Use Sequential Generation when you need consistent characters and scenes across variations
You can mix images from different sources — the model will adapt lighting and composition
For composition diversity, send results to the Carousel Generator
For complex edits, apply changes step by step
(e.g. first adjust pose, then modify clothing in a separate generation)

FAQ

Can I use it for NSFW images?
Yes. Models with minimal Safety Filters (such as General, WAN, and Flux Klein NSFW) support both SFW and NSFW content. Always follow platform rules.

What is the difference between Batch Mode and Sequential Generation?

Batch Mode processes each image separately using the same prompt
Sequential Generation creates a consistent sequence of images with shared characters and style

Can I change clothes or background only?
Yes. Describe the change clearly in the prompt (e.g. “change outfit to a black dress, keep face and pose”).
You can also upload a reference image — the model will use it during generation.

Will the editor preserve facial identity?
Yes, especially when:

a clear face image is provided
extreme or conflicting prompts are avoided

For maximum stability, avoid heavy stylization.
If needed, use Face Swap after generation.

Is this tool suitable for product or brand visuals?
Yes. For clean, brand-safe visuals, use Nano Banana 2 and keep prompts simple and descriptive.

Why didn’t I get the expected result?
This may happen if: