AI Image Generator (Image-to-Image)

AI-powered intelligent image editing tool supporting single and batch image editing, with optional masks for precise control

Bug report

Loading tool, please wait...

If the online tool fails to load or run, try the desktop tool.https://tools.yikeaigc.com/

Tool Usage

Return to old version

API Settings

API Domain

Enter your proxy API URL and key. This tool uses OpenAI's gpt-image-1 model, currently the most cost-effective option at about 2–4 cents per image. Check pricing with your proxy platform.

API Key

Your API key. Data is not saved; all operations are handled in your browser!

1Select Original Image

Supports JPG, PNG, WebP, etc., max 10MB

2Select Mask Image (Optional)

White areas will be edited; black areas stay unchanged. Leave empty to edit the whole image

3Edit Prompt

Describe the changes you want in detail

Example Edit Prompts

4Edit Settings

Output size

Output format

Compression Level (80%)

Edit Strength (0.7) Higher values make larger changes

1Select Original Image

Select multiple images in batch, max 10MB each

2Edit Prompt Template

Use {variable} format to define variables in the template

Sample templates

3Variable Settings

4Preview Generated Prompt

After setting a template and variables, the generated prompt preview will appear here

5Batch Parameters

Edit Mode

All Combinations

Sequential Edit

Unified Edit

Unified: use the same prompt for all images

Output size

Output format

Compression Level (80%)

Edit Strength (0.7)

Filename template Supported variables: {变量名}, {序号}, {原文件名}, {时间戳}

Concurrency (2) Image editing uses more resources; 1–5 recommended

Result Preview

Drag the center line

Slide left or right

Instructions

Software Usage Instructions

Configure API Information: Enter your API domain and key in the "API Configuration" area at the top. Supports the official OpenAI API and compatible third-party image editing API services.
Select an Editing Mode:
- Single Edit: Upload one image for editing
- Batch Edit: Upload multiple images and use templates for batch editing
Single Edit Mode:
- Upload the original image to edit (supports JPG, PNG, WebP, max 10MB)
- Optionally upload a mask image; white areas will be edited, while black areas remain unchanged
- Enter an editing prompt describing the desired modification effect
- Set the output size, format, compression level, and editing strength
Batch Edit Mode:
- Upload multiple images to edit
- Set an editing prompt template and use {变量名} to define replaceable parts
- Configure variable values; each variable can have multiple options
- Select the editing mode: all combinations, sequential editing, or unified editing
- Set a custom file name template and the number of concurrent processes
Start Editing: Click the corresponding edit button, and the tool will call the API to edit the image. You can view real-time progress.
Preview and Download: After editing is complete, preview the results in the results area. You can download files individually or package and download them in batches.

FAQ

A: Image editing modifies an existing image, such as changing the background, adding effects, or adjusting the style, while preserving the main elements of the original image; image generation creates a completely new image from scratch. Editing is better suited for optimizing existing materials, while generation is better suited for creating original content.

A: A mask is a black-and-white image. White areas indicate the parts to be edited, while black areas remain unchanged. You can create a mask using image editing software, or leave it blank to let the AI edit the entire image. A mask can precisely control the editing area and avoid accidentally changing important regions.

A: Full Combination Editing: each image is edited with all variable combinations; Sequential Editing: the 1st image uses the 1st row of variables, the 2nd image uses the 2nd row of variables; Uniform Editing: all images use the same editing prompt.

A: Edit strength controls how much the AI modifies the image. 0.1-0.3 is suitable for slight adjustments (such as color, lighting, and shadows); 0.4-0.7 is suitable for moderate changes (such as background and style); 0.8-1.0 is suitable for major changes. We recommend starting with 0.7 and adjusting based on the results.

A: Input supports common formats such as JPG, JPEG, PNG, WebP, and BMP, with each file no larger than 10MB. For output, you can choose JPEG (smaller file size), PNG (supports transparency), or WebP (balances quality and size).

A: Image editing needs to analyze the original image and make precise changes, which is more complex than generating from scratch. We recommend setting a lower concurrency level (1-3) to avoid processing too many images at the same time. Larger images take longer to process.

A: Effective prompts should be clear and specific: 1) describe the part you want to change (such as "background" or "color"); 2) state the desired result (such as "change to a sunset beach"); 3) avoid vague words and use specific descriptions; 4) you can specify a style (such as "cartoon style" or "oil painting effect").

A: AI editing regenerates the image, so the quality may differ slightly from the original. To maintain high quality: 1) use a high-quality original image; 2) choose an appropriate output size; 3) adjust the compression level; 4) avoid overly high editing strength. We recommend testing the effect on a small area first.