View images
A masonry grid for eyeballing every frame at once
- image
- ui_widget
- image
- image_list
ComfyUI's built-in image preview is fine for one image. It's awful for twenty. If you're doing batch generation, grid comparisons, or checking every frame of a video, you end up clicking through a thumbnail list that wasn't built for it. ViewImages is the LF Nodes answer: a masonry-style gallery widget that lays your images out in one scrollable view so you can actually compare them.
The mechanism is almost insultingly simple - feed it an image input (a single tensor or a list of them, the tooltip is explicit about both), and the KUL_MASONRY widget renders them in a grid. It's an output node, so it's a terminal in your graph. Crucially it also passes the images through: the outputs are image and image_list, both just passthroughs of whatever you fed in. You can put it in the middle of a graph and keep plumbing images downstream while using the grid purely as a monitor.
That passthrough is the part people miss, and it's why this node is better than a dead-end preview: wire a batch of images through it, keep the image_list going to a SaveImage or an upscaler, and you get a live comparison wall and a working pipeline.
What you set:
image- the one required input. Single tensor or a list; either works.
Where you'll genuinely reach for it: after a batch sampler spits out 8–16 variations and you want to pick winners; when you're testing a prompt across seeds and want the spread on screen at once; or mid-workflow as a debug view to confirm an image filter did what you thought before it goes on to the heavy nodes.
Installing LF Nodes (same pack for every LF node):
- ComfyUI Manager → search "LF Nodes" → install → restart, or
cd ComfyUI/custom_nodes && git clone https://github.com/lucafoscili/comfyui-lfthen restart.
Two things to know before you rely on it. First, the repo is in legacy mode - frozen at its last release while development moved to lucafoscili/lf-nodes - and this node carries over to the new repo under the same name. Second, the masonry widget is an LF/Ketchup Lite frontend, so it depends on the pack's web components loading correctly; if you've installed the pack manually, make sure the web directory shipped intact, because a broken frontend shows as a blank widget while the backend still happily passes images through. No models to download, and nothing in this node touches disk.
Inputs (2)
| Name | Type | Default | Description |
|---|---|---|---|
| image | IMAGE | Input image tensor or a list of image tensors. | |
| ui_widgetopt | KUL_MASONRY | [object Object] | — |
Outputs (2)
| Name | Type | Description |
|---|---|---|
| image | IMAGE | — |
| image_list | IMAGE | — |