Black Forest Labs | Flux Dev Lora
Generate images with Flux Dev model and custom LoRA weights from Replicate, HuggingFace, CivitAI, or direct URLs. Supports image-to-image mode and multiple output formats.
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": "black-forest-labs-flux-dev-lora",
"version": "0.0.1",
"input": {
"prompt": "a serene lake at twilight with reflections, photorealistic",
"lora_weights": "https://example.com/nature-lora.safetensors",
"lora_scale": 0.8,
"aspect_ratio": "16:9",
"guidance": 3
}
}'
Parameters
| Parameter |
Type |
Default |
Description |
| aspect_ratio |
string |
1:1 |
Aspect ratio of the output. enum: 1:1, 16:9, 21:9, 3:2, 2:3, 4:5, 5:4, 3:4, 4:3, 9:16, 9:21 |
| civitai_api_token |
string |
|
CivitAI API token for authenticated LoRA downloads. |
| disable_safety_checker |
boolean |
false |
Disable safety checker for generated images. |
| extra_lora |
string |
|
Additional LoRA weights URL. Supports Replicate and HuggingFace formats. |
| extra_lora_scale |
number |
1 |
Scale for the extra LoRA (0-2). |
| go_fast |
boolean |
false |
Run with fp8 quantized model for faster predictions. |
| guidance |
number |
3 |
Guidance scale for generated image. |
| hf_api_token |
string |
|
HuggingFace API token for authenticated LoRA downloads. |
| image |
string |
|
Input image for image-to-image mode. Output matches input aspect ratio. |
| lora_scale |
number |
1 |
Scale for the main LoRA (0-2). |
| lora_weights |
string |
|
Main LoRA weights. Supports Replicate, HuggingFace, CivitAI, or URL. |
| megapixels |
string |
1 |
Output megapixels. enum: 1, 0.25 |
| num_inference_steps |
integer |
28 |
Number of denoising steps (28-50 recommended). |
| num_outputs |
integer |
1 |
Number of outputs to generate. |
| output_format |
string |
webp |
Output format. enum: webp, jpg, png |
| output_quality |
integer |
80 |
Quality for output images (0-100). |
| prompt |
string |
|
Prompt for generated image. |
| prompt_strength |
number |
0.8 |
Prompt strength for img2img mode (1.0 = full destruction). |
| seed |
integer |
|
Random seed for reproducible generation. |
Examples
LoRA from HuggingFace:
curl -X POST https://api.eachlabs.ai/v1/prediction \
-H "Content-Type: application/json" \
-H "X-API-Key: $EACHLABS_API_KEY" \
-d '{
"model": "black-forest-labs-flux-dev-lora",
"version": "0.0.1",
"input": {
"prompt": "anime girl in a cherry blossom garden, detailed",
"lora_weights": "username/anime-style-lora",
"lora_scale": 0.9,
"aspect_ratio": "3:4",
"num_inference_steps": 30,
"output_format": "png"
}
}'
Image-to-image with dual LoRAs:
curl -X POST https://api.eachlabs.ai/v1/prediction \
-H "Content-Type: application/json" \
-H "X-API-Key: $EACHLABS_API_KEY" \
-d '{
"model": "black-forest-labs-flux-dev-lora",
"version": "0.0.1",
"input": {
"prompt": "vintage film photography style, warm tones",
"image": "https://example.com/input-photo.jpg",
"lora_weights": "https://example.com/vintage-lora.safetensors",
"lora_scale": 0.7,
"extra_lora": "https://example.com/film-grain-lora.safetensors",
"extra_lora_scale": 0.5,
"prompt_strength": 0.6,
"seed": 42
}
}'
Related Models
Documentation
1---2name: black-forest-labs-flux-dev-lora3description: Black Forest Labs | Flux Dev Lora. Generate images with Flux Dev and LoRA weights from multiple sources. Triggers: text to image, flux dev, lora, bfl, image generation, civitai, huggingface4---56# Black Forest Labs | Flux Dev Lora78Generate images with Flux Dev model and custom LoRA weights from Replicate, HuggingFace, CivitAI, or direct URLs. Supports image-to-image mode and multiple output formats.910## Quick Start1112> Requires an each::labs API key. Get one at [eachlabs.ai](https://eachlabs.ai).1314```bash15curl -X POST https://api.eachlabs.ai/v1/prediction \16 -H "Content-Type: application/json" \17 -H "X-API-Key: $EACHLABS_API_KEY" \18 -d '{19 "model": "black-forest-labs-flux-dev-lora",20 "version": "0.0.1",21 "input": {22 "prompt": "a serene lake at twilight with reflections, photorealistic",23 "lora_weights": "https://example.com/nature-lora.safetensors",24 "lora_scale": 0.8,25 "aspect_ratio": "16:9",26 "guidance": 327 }28 }'29```3031## Parameters3233| Parameter | Type | Default | Description |34|-----------|------|---------|-------------|35| aspect_ratio | string | 1:1 | Aspect ratio of the output. enum: 1:1, 16:9, 21:9, 3:2, 2:3, 4:5, 5:4, 3:4, 4:3, 9:16, 9:21 |36| civitai_api_token | string | | CivitAI API token for authenticated LoRA downloads. |37| disable_safety_checker | boolean | false | Disable safety checker for generated images. |38| extra_lora | string | | Additional LoRA weights URL. Supports Replicate and HuggingFace formats. |39| extra_lora_scale | number | 1 | Scale for the extra LoRA (0-2). |40| go_fast | boolean | false | Run with fp8 quantized model for faster predictions. |41| guidance | number | 3 | Guidance scale for generated image. |42| hf_api_token | string | | HuggingFace API token for authenticated LoRA downloads. |43| image | string | | Input image for image-to-image mode. Output matches input aspect ratio. |44| lora_scale | number | 1 | Scale for the main LoRA (0-2). |45| lora_weights | string | | Main LoRA weights. Supports Replicate, HuggingFace, CivitAI, or URL. |46| megapixels | string | 1 | Output megapixels. enum: 1, 0.25 |47| num_inference_steps | integer | 28 | Number of denoising steps (28-50 recommended). |48| num_outputs | integer | 1 | Number of outputs to generate. |49| output_format | string | webp | Output format. enum: webp, jpg, png |50| output_quality | integer | 80 | Quality for output images (0-100). |51| prompt | string | | Prompt for generated image. |52| prompt_strength | number | 0.8 | Prompt strength for img2img mode (1.0 = full destruction). |53| seed | integer | | Random seed for reproducible generation. |5455## Examples5657**LoRA from HuggingFace:**58```bash59curl -X POST https://api.eachlabs.ai/v1/prediction \60 -H "Content-Type: application/json" \61 -H "X-API-Key: $EACHLABS_API_KEY" \62 -d '{63 "model": "black-forest-labs-flux-dev-lora",64 "version": "0.0.1",65 "input": {66 "prompt": "anime girl in a cherry blossom garden, detailed",67 "lora_weights": "username/anime-style-lora",68 "lora_scale": 0.9,69 "aspect_ratio": "3:4",70 "num_inference_steps": 30,71 "output_format": "png"72 }73 }'74```7576**Image-to-image with dual LoRAs:**77```bash78curl -X POST https://api.eachlabs.ai/v1/prediction \79 -H "Content-Type: application/json" \80 -H "X-API-Key: $EACHLABS_API_KEY" \81 -d '{82 "model": "black-forest-labs-flux-dev-lora",83 "version": "0.0.1",84 "input": {85 "prompt": "vintage film photography style, warm tones",86 "image": "https://example.com/input-photo.jpg",87 "lora_weights": "https://example.com/vintage-lora.safetensors",88 "lora_scale": 0.7,89 "extra_lora": "https://example.com/film-grain-lora.safetensors",90 "extra_lora_scale": 0.5,91 "prompt_strength": 0.6,92 "seed": 4293 }94 }'95```9697## Related Models9899- [BFL Flux Lora](../bfl-flux-lora/) - BFL Flux with LoRA support100- [Flux 2 | Text to Image Lora](../flux-2-lora/) - Flux 2 with LoRA support101102## Documentation103104- [each::labs Docs](https://docs.eachlabs.ai)105- [API Reference](https://docs.eachlabs.ai/api/overview)
awesome-genmedia