Any Passthrough (1 → 6)

Any Passthrough (1 → 6) is a fan-out node: one input in, six identical outputs out. You'd think ComfyUI lets you drag the same wire to as many nodes as you want, and you'd be right - so at first glance this looks redundant. Where it earns its place is the fan-out that's conditional or reroutable: it gives you a single stable point that feeds six downstream sockets, so you can rewire the whole branch by moving one connection instead of six. In big graphs it also keeps things readable, which is worth more than it sounds.

How it works

Mechanically it's the simplest node in the pack: the input uses ComfyUI's match-type system, so whatever type you connect - IMAGE, LATENT, CONDITIONING, a STRING, anything - is preserved across all six outputs (out1 through out6), and the single input value is echoed to each. No state, no transformation, no side effects. That type preservation is the whole trick: because the output ports carry the connected type, downstream nodes connect cleanly without you reaching for an "any" socket and hoping.

Input:

• input - any single value. Outputs: out1 … out6, all identical.

The natural pairing is with the pack's other passthrough, Any Passthrough (6 → 1), which does the reverse (first non-null of six inputs becomes one output). Together they give you the two primitive routing moves - split one into many, merge many into one - which is exactly the kind of plumbing that makes a modular workflow easy to rearrange.

Installing

This is one of the 26 nodes in ComfyUI-Flow-Assistor - ComfyUI Manager (search "Flow Assistor") or:

cd ComfyUI/custom_nodes
git clone https://github.com/Merserk/ComfyUI-Flow-Assistor.git

Restart after cloning. Current ComfyUI required (the pack is V3-only); no extra dependencies.

Where people get burned

The only real surprise is how dumb it is - people expect a fan-out node to do something (decorate, sync, filter) and it does none of it. Don't use it as a way to duplicate work, because it isn't: all six outputs are the same Python object, and if that object is mutable, downstream nodes see the same underlying data, not six copies. And if you only need two or three fan-outs and you're optimizing for fewer nodes, this is more machinery than you need - it's a readability and rerouting tool, not a performance one. Where it genuinely shines is a workflow you intend to remix: swap the input once and every downstream consumer follows.