Edit any image with a custom-trained LoRA style applied on top of Qwen Image Edit. Precise, fine-tuned edits via a simple REST API. Try free — pay per generation, no subscription.
Qwen Image Edit LoRA is a fine-tuned image editing model with custom LoRA support. Modifying or adding elements with fine-tuned styles.
About this model
Qwen Image Edit LoRA is Qwen Image Edit with custom LoRA support layered on top, letting you apply a fine-tuned style, character, or product look while modifying or adding elements to an input image. Provide a single source image, a text prompt describing the edit, and up to 3 LoRA weight files with individual scale multipliers.
Cost analysis
| Provider | Cost | Notes |
|---|---|---|
| muapiapp | $0.04 per generation | Flat per-generation price regardless of how many LoRA weights are applied. |
| Fal.ai | $0.05 per generation | Muapiapp is roughly 20% cheaper for comparable Qwen Image Edit LoRA generations. |
| Replicate | $0.05 per generation | Usage-based pricing comparable to Fal.ai; muapiapp remains the cheaper option. |
Flat per-generation price regardless of how many LoRA weights are applied.
Muapiapp is roughly 20% cheaper for comparable Qwen Image Edit LoRA generations.
Usage-based pricing comparable to Fal.ai; muapiapp remains the cheaper option.
** Competitor pricing is estimated based on similar model architectures and usage tiers.
Configuration schema
| Parameter | Type | Description | Default |
|---|---|---|---|
| Prompt | string | Text prompt describing the image edits. | A cinematic ocean wave at sunrise, highly detailed |
| Image URL | string | URL of the input image. | https://d3adwkbyhxyrtq.cloudfront.net/ai-images/186/902675646946/c0951838-8bc5-4598-8e8e-941df16446fa.jpg |
| Aspect Ratio | Enum (9 options) | Aspect ratio of generated image. | 1:1 |
| LoRAs | array | List of LoRAs to apply (maximum 3). | [object Object] |
| Output Format | Enum (3 options) | Format of the output image. | jpeg |
Text prompt describing the image edits.
A cinematic ocean wave at sunrise, highly detailedURL of the input image.
https://d3adwkbyhxyrtq.cloudfront.net/ai-images/186/902675646946/c0951838-8bc5-4598-8e8e-941df16446fa.jpgAspect ratio of generated image.
1:1List of LoRAs to apply (maximum 3).
[object Object]Format of the output image.
jpegDeveloper documentation
Provide the Source Image
image_url to the image you want to edit.Write the Edit Prompt
Apply LoRA Weights (Optional)
loras, each with a path (URL to a .safetensors file) and a scale (0–4, default 1) controlling its strength.Set Output Options
aspect_ratio and output_format (jpeg, png, or webp).Submit via the API
curl -X POST https://api.muapi.ai/api/v1/qwen-image-edit-lora \
-H "x-api-key: YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"prompt": "Add a red scarf and change the background to a snowy street.",
"image_url": "https://example.com/source.jpg",
"loras": [{"path": "https://example.com/my-style.safetensors", "scale": 1}]
}'
GET /api/v1/predictions/{request_id}/result until status is completed, then retrieve the image URL from output.Frequently asked
It edits a single input image according to your text prompt, while applying up to 3 custom LoRA weight files to steer the result toward a trained style, character, or look.
Any publicly reachable `.safetensors` URL — including one produced by a muapiapp LoRA trainer endpoint, or a third-party host like Hugging Face.
Up to 3, each with its own `scale` multiplier (0–4, default 1) to control how strongly it influences the output.
This endpoint takes a single `image_url`. The 2511 variant accepts up to 3 reference images via `images_list` and offers stronger multi-image edit consistency.
`jpeg`, `png`, or `webp`, set via `output_format`.