EasyControlnet(controlnetSimple)

Description

The EasyControlnet node (controlnetSimple) is a loader/application node in the EasyUse/Loaders category. It takes an existing pipeline and a control image, applies a ControlNet model to the pipeline's conditioning, and returns the updated pipeline plus positive and negative conditioning outputs.

This node can be used in two ways: select a ControlNet by name from the control_net_name dropdown, or connect a loaded CONTROL_NET object directly to the optional control_net input.

Inputs

pipe

A PIPE_LINE containing the working model, positive and negative conditionings, latents, and other pipeline state. The ControlNet is applied to this pipeline's conditioning, and the updated pipeline is returned as the first output.

image

An IMAGE tensor used as the ControlNet conditioning image. This should match whatever the selected ControlNet expects, such as an edge map, pose skeleton, depth map, or other structural guide.

control_net_name

A required combo box listing the available ControlNet models. Choose the model to load by name. The available entries come from the ControlNet models installed in your environment.

control_net

Optional CONTROL_NET input. If a loaded ControlNet object is connected here, it is used in place of loading the model by name.

strength

Float, default 1.0, range 0.0 to 10.0, step 0.01.

Controls how strongly the ControlNet affects the conditioning. A value of 0.0 disables its influence; higher values increasingly enforce the control image's structure.

scale_soft_weights

Float, default 1.0, range 0.0 to 1.0, step 0.001.

Scales the soft weighting of the ControlNet. Lower values reduce the per-block/soft weight contribution, while 1.0 applies the full soft-weight scale.

Outputs

pipe output

The updated PIPE_LINE with the ControlNet-applied conditioning included, ready to be passed to downstream pipeline nodes.

conditioning (positive)

Positive conditioning with the ControlNet applied. Connect this to the positive conditioning input of a sampler.

conditioning (negative)

Negative conditioning with the ControlNet applied. Connect this to the negative conditioning input of a sampler.

The output order is: updated PIPE_LINE, positive conditioning, then negative conditioning.

Usage Notes

• Use both CONDITIONING outputs with a KSampler or compatible sampler node.
• If you need to apply multiple ControlNets, chain additional EasyControlnet nodes or combine the conditionings before sampling.
• The control_net_name dropdown is required by the node UI; when an explicit control_net object is connected, it supplies the model directly.
• Start with strength at 1.0 and adjust based on how strongly the control image should drive the result.
• Start with scale_soft_weights at 1.0; lower it only if the ControlNet is over-contributing or producing artifacts.
• The control image should generally have the same dimensions as the generated image or be resized consistently by the model pipeline.