GR Image Details Displayer

This one's not for wiring further downstream - it's a debug/inspection node. It takes an images input and nothing else, produces no data outputs at all, and is explicitly marked as an output node in its own schema. That combination means its whole job is to display information about whatever image batch you feed it, right there in the UI, rather than passing anything along the graph.

How it works

Because ComfyUI lets a single output feed more than one downstream node, you can splice this in without breaking your actual pipeline: take the same images wire that's already going into your save node or your next processing step, and also connect it here. It doesn't consume or alter the data - it just gives you a place to look at what's actually flowing through that point in the graph, which is genuinely useful when you're debugging a workflow and want to confirm a batch has the dimensions, count, or shape you think it does before it hits an expensive downstream step.

The schema doesn't specify exactly what gets displayed - that's determined by the node's own UI rendering rather than anything visible in its inputs or outputs - so the honest answer is: connect it, run your graph, and look at what shows up rather than assuming a specific set of fields ahead of time.

The inputs and outputs that matter

• images - the only input, required. No optional inputs, no configuration knobs.
• No outputs. This is a dead-end node by design - it's meant to be a branch off your main pipeline, not a link in the chain.

How to install it

ComfyUI Manager, search GraftingRayman. Manual install:

cd ComfyUI/custom_nodes
git clone https://github.com/GraftingRayman/ComfyUI_GraftingRayman

Restart afterward. This pack requires OpenAI's CLIP installed separately or nothing imports, and that applies here too even though this node does nothing with vision models - the whole pack shares one Python module and one failed import takes it all down. Portable ComfyUI: .\python_embeded\python.exe -m pip install git+https://github.com/openai/CLIP.git. System Python: pip install git+https://github.com/openai/CLIP.git.

Common issues & troubleshooting

Pack fails to load. Check your startup console for the CLIP import error and run the pip install above.

Nothing seems to happen when you run the graph. Since this node has no output, there's nothing to trace downstream - the information it produces shows up in the node itself or the console, not as data flowing further into your graph. If you're expecting a value to wire elsewhere, that's not what this node is for; it's read the value where it's displayed, not pull it into another node.

Connecting it seems to slow down or duplicate work in your graph. It shouldn't - ComfyUI's execution graph reuses a shared upstream output for multiple downstream consumers rather than recomputing it, so branching your images wire into this node alongside your real pipeline shouldn't cost you an extra generation pass. If you do see a slowdown, look elsewhere in the graph first.

You want this information available programmatically, not just displayed. This node is a UI-facing inspector with no data output, so it's not the right fit if you need image metadata to feed another node - you'd want a node that actually returns values (dimensions, batch count) as outputs instead.