Qwen 3.0 Image to Image API: AI Image Editor

Transform and edit reference images using Qwen 3.0 Image to Image API on Muapi. High-precision background, element, and poster content replacement.

📝

Overview

About this model

Qwen 3.0 Image to Image transforms existing reference photographs according to text editing instructions while preserving ambient lighting, composition, perspective, and fine environment details.

1Targeted Poster & Billboard Editing: Replace poster or sign artwork inside complex indoor/outdoor photographs.

2Virtual Restyling & Retouching: Change clothing textures, wall patterns, or individual objects while keeping the surrounding room intact.

3Multi-Reference Compositing: Combine visual elements from reference images into unified new scenes.

💰

Pricing & Value

Cost analysis

Provider	Cost	Notes
muapiapp	$0.03 (1K) / $0.05 (2K)	Includes multi-image input processing.
Fal.ai	$0.04 (1K) / $0.06 (2K)	muapiapp remains more cost-effective at every resolution tier.
Replicate	$0.04 (1K) / $0.065 (2K)	muapiapp remains more cost-effective at every resolution tier.

muapiapp$0.03 (1K) / $0.05 (2K)

Includes multi-image input processing.

Fal.ai$0.04 (1K) / $0.06 (2K)

muapiapp remains more cost-effective at every resolution tier.

Replicate$0.04 (1K) / $0.065 (2K)

muapiapp remains more cost-effective at every resolution tier.

* Competitor pricing is estimated based on similar model architectures and usage tiers.

⚙️

Technical Details

Configuration schema

Parameter	Type	Description	Default
Prompt	string	The positive prompt describing image transformation, style, and editing instructions. (Max 800 characters)	`Preserve the entire coffee shop environment, furniture, lighting, and camera perspective. Replace only the content of the framed poster.`
Input Images	array	Input reference image URLs (up to 10MB each).	`https://static.aiquickdraw.com/tools/example/1785909232707_STAQCAqg.png`
Resolution	Enum (2 options)	-	`1k`
Aspect Ratio	Enum (8 options)	-	`16:9`
Output Format	Enum (2 options)	-	`png`
Intelligent Prompt Extend	boolean	-	`true`
Negative Prompt	string	Described content to exclude from the generated image. (Max 5000 characters)	`-`

Promptstring

The positive prompt describing image transformation, style, and editing instructions. (Max 800 characters)

Default Value

Preserve the entire coffee shop environment, furniture, lighting, and camera perspective. Replace only the content of the framed poster.

Input Imagesarray

Input reference image URLs (up to 10MB each).

Default Valuehttps://static.aiquickdraw.com/tools/example/1785909232707_STAQCAqg.png

ResolutionEnum (2 options)

Default Value1k

Aspect RatioEnum (8 options)

Default Value16:9

Output FormatEnum (2 options)

Default Valuepng

Intelligent Prompt Extendboolean

Default Valuetrue

Negative Promptstring

Described content to exclude from the generated image. (Max 5000 characters)

Default Value-

📖

Implementation Guide

Developer documentation

How to Use Qwen 3.0 Image to Image API

Provide reference images: Pass array of image URLs in image_urls (up to 10MB each).
Write editing instructions: Describe what to preserve and what specific elements to replace or modify.
Select output format & resolution: Set resolution (1K or 2K) and image_size aspect ratio.
Submit task: Send a POST request to /api/v1/qwen3-image-to-image.

curl -X POST "https://api.muapi.ai/api/v1/qwen3-image-to-image" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "image_urls": ["https://example.com/reference.png"],
    "prompt": "Preserve the coffee shop environment and replace only the framed poster content.",
    "resolution": "1K"
  }'

❓

Common Questions

Frequently asked

What is Qwen 3.0 Image to Image?

Qwen 3.0 Image to Image is an advanced image-to-image translation model capable of precise local edits and style transformations.

How many reference images can I upload?

You can provide multiple reference image URLs (up to 10MB per image file).

Does it preserve lighting and camera perspective?

Yes, Qwen 3.0 Image to Image maintains physical lighting, reflections, and camera perspectives from the input image.