Skip to main content
This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub
The LoraModelLoader node applies trained LoRA (Low-Rank Adaptation) weights to a diffusion model. It modifies the base model by loading LoRA weights from a trained LoRA model and adjusting their influence strength. This allows you to customize the behavior of diffusion models without retraining them from scratch.

Inputs

ParameterData TypeRequiredRangeDescription
modelMODELYes-The diffusion model the LoRA will be applied to.
loraLORA_MODELYes-The LoRA model to apply to the diffusion model.
strength_modelFLOATYes-100.0 to 100.0How strongly to modify the diffusion model. This value can be negative (default: 1.0).
Note: When strength_model is set to 0, the node returns the original model without applying any LoRA modifications.

Outputs

Output NameData TypeDescription
modelMODELThe modified diffusion model with LoRA weights applied.