Lora Info (LoraManager)

This is a look-but-don't-touch node. It shows you information - the filename and any notes you've saved - about a LoRA you've selected, right there on the canvas, without affecting a single thing your workflow produces. If you've ever right-clicked a LoRA in rgthree's Power Lora Loader to read its trigger words and notes, this is that idea as a standalone display node.

Why bother? Because in a busy graph, the LoRAs you loaded are buried inside a text box or a stack, and it's easy to lose track of which file is which - especially when your collection is full of cryptically-named .safetensors downloads. Dropping a Lora Info node next to your loader gives you a live readout of what you're actually pulling in, including any usage notes you jotted down in the manager UI. It's the in-canvas version of flipping back to the model page.

How it works

Per the node's own description: connect any output from a Lora Loader or a Stacker into the Info node's source input, then pick a LoRA in the source widget - the info panel updates automatically to show that LoRA's filename and notes. It reads, it displays, and it stops there. It does not modify your model, your conditioning, or anything downstream; it's purely a viewer.

Inputs and outputs that matter

Here's the honest bit: this node has no standard typed inputs or outputs in its schema. It's a display widget, not a data-processing node - the connection to a loader or stacker is a UI hook, and the output is the panel you read, not a wire you route onward. So there's nothing to plug into a sampler here. You wire into it for context; you don't wire out of it.

That's normal for inspection nodes. Think of it like a label or a note that happens to pull live data from the manager.

How to install it

Via ComfyUI Manager: search lora-manager, Install, restart. Or by hand:

cd ComfyUI/custom_nodes
git clone https://github.com/willmiao/ComfyUI-Lora-Manager.git
cd ComfyUI-Lora-Manager
pip install -r requirements.txt

then restart. Light Python deps, no models to download.

Common issues & troubleshooting

No notes show up. Notes are something you add yourself in the LoRA Manager web UI (the "custom notes and usage tips" feature). A fresh LoRA you never annotated will show its filename and not much else - that's expected, not a bug.

It's not changing my output. Correct, and by design. This node is deliberately inert; it exists to inform, not to process. If you wanted the trigger words to actually reach your prompt, that's the loader/stacker's trigger_words output, not this node.

The panel is stale. Re-select the LoRA in the source widget to refresh it, and make sure the manager has current metadata for that file - if you renamed or re-downloaded the LoRA, rescan in the UI.