ER Wedge Preview Image

ER Wedge Preview Image is the node you drop in when the wedges are done and the results have to leave your machine. The ER Wedge Sequencer gives you a working grid inside the node - perfect for you, but nobody else wants to look at your graph. This is the client-facing version: a clean gallery that mirrors every image your wedge sequencers produced, grouped per wedge and labelled with the parameter values, ready to present or export.

It's part of the ER VFX Nodepack by Héctor Gallego (ER Academy), and it's deliberately boring to set up. In fact it's about the only node you'll ever add that needs no wiring at all: no inputs, no outputs. It's a front-end-only node - the Python side is literally a no-op - that watches the graph for ER Wedge Sequencer results and mirrors them. That's the whole trick, and it's why it can sit anywhere in the workflow without touching your data flow.

What you get

Each sequencer's results appear grouped under its own heading - WEDGE 1, WEDGE 2, and so on - with the values that produced each image overlaid or listed per thumbnail, so a three-way seed comparison reads as "these three frames, seeds 7/77/777" at a glance rather than as cryptic thumbnails. If you have more than one sequencer in the graph, pick which one to show from the dropdown; otherwise it just mirrors everything automatically.

Two things it's genuinely good for:

• Sending a look to a client or collaborator without handing them a ComfyUI workflow or ten loose files. Type a custom title and use the 🖼 Contact sheet button (or right-click) to export one PNG containing every result with its labels - the standard "here are your options" deliverable in VFX look-dev.
• Documenting your own experiments. A contact sheet is a better record than a folder of screenshots because the values are baked into the image.

What it isn't

It doesn't run wedges - that's the sequencer's job. It doesn't process pixels or pass data along (no sockets at all). And it only sees results produced by ER Wedge Sequencer nodes in the current graph, not arbitrary images. If you're wedging video, grab the sibling ER Wedge Preview Video instead, which does the same gallery for frame-batch results with autoplaying clips.

Installing

Ships inside the ER VFX Nodepack. ComfyUI Manager → search "ER VFX Nodepack" → Install, or:

cd ComfyUI/custom_nodes
git clone https://github.com/esperandoelrender/ComfyUI-ER-VFX-Nodepack.git

Restart and find it under utils (search "ER Wedge"). No dependencies, no model downloads. If you previously had the standalone er_* packs installed, remove those folders - this pack supersedes them and duplicates would collide.

One small note: since this node only renders what's already in the graph, the wedge runs have to be done first. Drop it in before you start wedging if you want to watch the gallery fill up live, or after - either works, it just mirrors current results.