FLUX.1 [dev] with Controlnets and Loras

A specialized FLUX endpoint combining differential diffusion control with LoRA, ControlNet, and IP-Adapter support, enabling precise, region-specific image transformations through customizable change maps.

Input

Fields marked * are required. Review the price before generating.

Prompt *

The prompt to generate an image from.

Image Size

The size of the generated image.

Num Images

The number of images to generate. This is always set to 1 for streaming output.

Reference Image

URL of Image for Reference-Only

Uploading a file needs an account. Create accountJPEG, PNG, WebP · up to 25 MB

Image *

URL of image to use as initial image.

Uploading a file needs an account. Create accountJPEG, PNG, WebP · up to 25 MB

Change Map URL *

URL of change map.

Uploading a file needs an account. Create accountJPEG, PNG, WebP · up to 25 MB

Additional settings

Num Inference Steps

The number of inference steps to perform.

Seed

The same seed and the same prompt given to the same version of the model will output the same image every time.

Loras

The LoRAs to use for the image generation. You can use any number of LoRAs and they will be merged together to generate the final image.

Control Loras

Tips and requirements

The LoRAs to use for the image generation which use a control image. You can use any number of LoRAs and they will be merged together to generate the final image.

Controlnets

The controlnets to use for the image generation. Only one controlnet is supported at the moment.

Controlnet Unions

The controlnet unions to use for the image generation. Only one controlnet is supported at the moment.

Ip Adapters

IP-Adapter to use for image generation.

Easycontrols

EasyControl Inputs to use for image generation.

ImageFillInput

Use an image input to influence the generation. Can be used to fill images in masked areas.

Fill Image

URLs of images to be filled for redux prompting

Uploading a file needs an account. Create accountJPEG, PNG, WebP · up to 25 MB

Guidance scale (CFG)

Tips and requirements

The CFG (Classifier Free Guidance) scale is a measure of how close you want the model to stick to your prompt when looking for a related image to show you.

Real CFG scale

Tips and requirements

The CFG (Classifier Free Guidance) scale is a measure of how close you want the model to stick to your prompt when looking for a related image to show you.

Use Real CFG

Tips and requirements

Uses classical CFG as in SD1.5, SDXL, etc. Increases generation times and price when set to be true. If using XLabs IP-Adapter v1, this will be turned on!.

Use CFG-Zero-Init

Uses CFG-zero init sampling as in https://arxiv.org/abs/2503.18886.

Reference Strength

Strength of reference_only generation. Only used if a reference image is provided.

Reference Start

The percentage of the total timesteps when the reference guidance is to bestarted.

Reference End

The percentage of the total timesteps when the reference guidance is to be ended.

Base Shift

Base shift for the scheduled timesteps

Max Shift

Max shift for the scheduled timesteps

Output Format

The format of the generated image.

Use Beta Schedule

Specifies whether beta sigmas ought to be used.

Sigma Schedule

Sigmas schedule for the denoising process.

Scheduler

Scheduler for the denoising process.

Negative Prompt

Tips and requirements

Negative prompt to steer the image generation away from unwanted features. By default, we will be using NAG for processing the negative prompt.

NAG scale

The scale for NAG. Higher values will result in a image that is more distant to the negative prompt.

NAG Tau

Tips and requirements

The tau for NAG. Controls the normalization of the hidden state. Higher values will result in a less aggressive normalization, but may also lead to unexpected changes with respect to the original image. Not recommended to change this value.

NAG alpha

Tips and requirements

The alpha value for NAG. This value is used as a final weighting factor for steering the normalized guidance (positive and negative prompts) in the direction of the positive prompt. Higher values will result in less steering on the normalized guidance where lower values will result in considering the positive prompt guidance more.

Proportion of steps to apply NAG

Tips and requirements

The proportion of steps to apply NAG. After the specified proportion of steps has been iterated, the remaining steps will use original attention processors in FLUX.

Strength

The strength to use for differential diffusion. 1.0 is completely remakes the image while 0.0 preserves the original.

Price$0.60 / run

You sign up first. Your settings are kept.

API code for this input

Complete the required fields and fix input errors to see API code.

How to use FLUX.1 [dev] with Controlnets and Loras

Getting started

Apply spatially guided changes when different parts of the source need different treatment.

Review your result

Inspect transitions around edited regions and compare areas meant to retain their original appearance.

Inputs and settings

Provide image_url, change_map_image_url and prompt. Start without extra adapters to see the map’s effect clearly.

Tune strength within 0.01–1. Add compatible controlnets or loras separately, keeping seed fixed while comparing each change.

Input

How to use FLUX.1 [dev] with Controlnets and Loras

Getting started

Review your result

Inputs and settings

Related models