Show Loader Settings Names

Simple, honest utility node: it looks inside a pipe and tells you, as plain text, which checkpoint, VAE, and LoRA actually got loaded. That sounds trivial until you're several nodes deep in a workflow where the pipe passed through two or three loaders and an edit node, and you genuinely can't tell at a glance what model is live anymore.

Why this exists. Easy-Use's loaders bundle everything into a pipe, which is great for keeping your graph tidy but bad for at-a-glance visibility - a PIPE_LINE wire doesn't show you what's inside it the way a labeled CheckpointLoader node does. easy showLoaderSettingsNames reads that pipe and surfaces the model, VAE, and LoRA names it's carrying as actual strings, so you can display them, log them, or feed them somewhere that needs to know what generated a given image.

Inputs and outputs. One required input, pipe. Three outputs, all STRING: ckpt_name, vae_name, lora_name. It's marked an output node in its own right, so on its own it'll display those names directly on the node - but because the outputs are typed sockets, you can also wire them onward. A common pattern: feed ckpt_name into a text-overlay node so your saved images get watermarked with which checkpoint produced them, or concatenate all three into a filename so your output folder is self-documenting without you tracking it by hand.

It's not a substitute for the workflow's own metadata. ComfyUI already embeds the full workflow JSON into a saved PNG's metadata - technically, everything this node reports is recoverable by dragging the image back into ComfyUI and reading the graph. What this node buys you is a human-readable answer without doing that round-trip: someone looking at a stack of output images, or reading a log, doesn't want to reload a workflow JSON just to confirm which checkpoint made which batch. Pairing this with easy string (to build a filename) or easy tableEditor (to log a run's settings as a readable table) is a natural combination if you're doing anything approaching batch comparisons across checkpoints or LoRAs.

Install. ComfyUI Manager, search "ComfyUI Easy Use," or manually:

cd ComfyUI/custom_nodes
git clone https://github.com/yolain/ComfyUI-Easy-Use

then restart. No extra dependencies - it's reading metadata off the pipe object, not touching models or files on disk.

Common issues. The main gotcha is what happens when a slot is genuinely empty - if the pipe you're reading from never had a LoRA loaded (no LoRA in the stack), lora_name will come back blank rather than throwing an error, which is correct behavior but can look like the node is broken if you were expecting a name. Also worth knowing: this reads whatever the pipe currently says, not some persistent generation record - if you passed the pipe through easy pipeEdit and swapped the model partway through your graph, this node reports the swapped-in name, not the original. That's usually exactly what you want (you're asking "what's actually loaded right now"), but it's worth remembering if you're trying to audit the very first loader in a long chain.