awesome-genmedia Flux 2 | Lora Edit
Edit images using up to 3 LoRA weights for specialized style control. Combines Flux 2's editing capabilities with LoRA-based fine-tuning for targeted modifications.
Quick Start
Requires an each::labs API key. Get one at eachlabs.ai.
curl -X POST https://api.eachlabs.ai/v1/prediction \
-H "Content-Type: application/json" \
-H "X-API-Key: $EACHLABS_API_KEY" \
-d '{
"model": "flux-2-lora-edit",
"version": "0.0.1",
"input": {
"prompt": "transform into a vintage film photograph style",
"image_urls": ["https://example.com/photo.jpg"],
"loras": [{"path": "https://example.com/vintage-lora.safetensors", "scale": 0.8}],
"guidance_scale": 2.5,
"image_size": "square_hd"
}
}'
Parameters
| Parameter | Type | Default | Description |
|---|---|---|---|
| acceleration | string | regular | The acceleration level to use. enum: none, regular, high |
| enable_prompt_expansion | boolean | false | If set to true, the prompt will be expanded for better results. |
| enable_safety_checker | boolean | true | If set to true, the safety checker will be enabled. |
| guidance_scale | number | 2.5 | Guidance Scale controls how closely the model follows your prompt. |
| image_size | string | square_hd | The size of the image to generate. enum: square_hd, square, portrait_4_3, portrait_16_9, landscape_4_3, landscape_16_9 |
| image_urls | array | The URLs of the images for editing. | |
| loras | array | List of LoRA weights to apply (maximum 3). | |
| num_images | integer | 1 | The number of images to generate. |
| num_inference_steps | integer | 28 | The number of inference steps to perform. |
| output_format | string | png | The format of the generated image. enum: jpeg, png, webp |
| prompt | string | The prompt to generate an image from. | |
| seed | integer | The seed to use for the generation. |
Examples
LoRA-guided style edit:
curl -X POST https://api.eachlabs.ai/v1/prediction \
-H "Content-Type: application/json" \
-H "X-API-Key: $EACHLABS_API_KEY" \
-d '{
"model": "flux-2-lora-edit",
"version": "0.0.1",
"input": {
"prompt": "apply anime illustration style to the portrait",
"image_urls": ["https://example.com/portrait.jpg"],
"loras": [{"path": "https://example.com/anime-lora.safetensors", "scale": 0.9}],
"guidance_scale": 3.0,
"num_inference_steps": 30,
"output_format": "png"
}
}'
Multi-LoRA editing with acceleration:
curl -X POST https://api.eachlabs.ai/v1/prediction \
-H "Content-Type: application/json" \
-H "X-API-Key: $EACHLABS_API_KEY" \
-d '{
"model": "flux-2-lora-edit",
"version": "0.0.1",
"input": {
"prompt": "enhance with cinematic lighting and film grain",
"image_urls": ["https://example.com/scene.jpg"],
"loras": [
{"path": "https://example.com/cinematic-lora.safetensors", "scale": 0.7},
{"path": "https://example.com/film-grain-lora.safetensors", "scale": 0.5}
],
"acceleration": "high",
"num_images": 2,
"seed": 42
}
}'
Related Models
- Flux 2 | Text to Image Lora - LoRA-based text-to-image generation
- Flux 2 | Edit - Standard image editing without LoRA