Patch Model Patcher Order(PatchModelPatcherOrder)
The technical core of this node is to modify the execution order of the internal `patch_model` function in ComfyUI.
Patch Model Patcher Order
The Patch Model Patcher Order node. This node is deprecated and no longer functional. It is kept in the codebase only for backwards compatibility. Any previous functionality – modifying the execution order of the internal patch_model function in ComfyUI – has been removed. Use TorchCompileModelAdvanced to apply LoRA with torch.compile.
Node Functionality
This node no longer performs any operation. It exists solely to prevent breakage in existing workflows that reference it. The original purpose was to adjust the loading order of model patches (object patches first or weight patches first) and control full model loading, but those features are now obsolete.
Node Parameter Description - Patch Model Patcher Order
Connection Inputs (Inputs)
| Parameter Name | Data Type | Required | Default Value | Value Range/Options | Description |
|---|---|---|---|---|---|
model | MODEL | Yes | - | - | **: Receives the input model to be processed, serving as the baseline for subsequent operations. |
Control Parameters (Parameters)
| Parameter Name | Data Type | Required | Default Value | Value Range/Options | Description |
|---|---|---|---|---|---|
patch_order | COMBO | Yes | weight_patch_first | "object_patch_first", "weight_patch_first" | **: Selects the loading order for model patches. Available options are object_patch_first (object patches first) or weight_patch_first (weight patches first). |
full_load | COMBO | Yes | auto | "enabled", "disabled", "auto" | disabled (disabled) or auto (automatic). Disabling may help alleviate memory issues when loading large models. |
Output
| Parameter Name | Data Type | Description |
|---|---|---|
| MODEL | MODEL | **: Outputs the model after (theoretically) adjusting the patch order. |
Usage Scenarios
This node should not be used in new workflows. It is retained only for backward compatibility and does nothing. If you need to use LoRA with torch.compile, replace instances of this node with TorchCompileModelAdvanced.
Notes
Please note that this node has been marked as deprecated and no longer has actual functionality. The source code explicitly states that its functionality has been superseded by the more advanced TorchCompileModelAdvanced node. This node is retained only for backward compatibility and should not be used in new workflows.
Patch Model Patcher Order Node Source Code Link
The Patch Model Patcher Order node is from the ComfyUI-KJNodes node package.
Comments
Sign in with GitHub to join the discussion.