Skip to content

Wiring & Fixing Load LoRA (LoraLoader)

Practical pin-by-pin connection order and common error fixes for Load LoRA (LoraLoader) in ComfyUI workflows.

Updated Oct 17, 2026 · 1 min read

At a glanceHow-to
Target Node
Load LoRA (LoraLoader)
Required Inputs
model, clip, lora_name, strength_model, strength_clip
Output Pins
MODEL, CLIP
Package
Built-in Core

Step-by-step pin wiring for Load LoRA (LoraLoader)

  1. Drop Load LoRA (LoraLoader) onto the canvas by double-clicking and searching its title.
  2. Connect all required upstream wires (`model, clip, lora_name, strength_model, strength_clip`) from your checkpoint loader, prompt conditioning, or image preprocessor.
  3. Route the output (`MODEL, CLIP`) into your downstream KSampler or VAE Decode node, then click **Queue Prompt**.

Fixing dependency conflicts and type mismatches

If the console prints a Python import error or tensor shape mismatch on this node, verify that your checkpoint family (SD 1.5, SDXL, or Flux) matches any attached LoRA, ControlNet, or IPAdapter weight.

For custom node packs on Windows, run **Update All** inside ComfyUI-Manager or grab a fresh desktop environment from /download.

FAQs

Why won't a wire snap onto Load LoRA (LoraLoader)?

ComfyUI enforces strict pin types. Check that the upstream output matches `model, clip, lora_name, strength_model, strength_clip` and insert a VAE Encode/Decode or preprocessor node if converting between pixels and latents.

How do I bypass this node temporarily without deleting my wires?

Select the node on the canvas and press **Ctrl+B** (Bypass) or **Ctrl+M** (Mute) to test your graph without disconnecting cables.

Run on your hardware. Free forever.

The full ComfyUI engine: open source, fast, extensible, and yours to run however you want. Download the desktop app and open this workflow on your own machine.

Sofa split between tweed and snow-leopard fabric, an image edit made in ComfyUI