8 Floats

8 Floats is the float twin of Mira's 8 Bools node. It gives you eight number fields in one node and packs them into a single FLOAT_LIST output, so you can carry a set of values down one wire and unpack the ones you need with a Float List Interpreter at the other end. If you find yourself running eight separate float links across a big graph, this collapses them into one.

It's a Numeral node in ComfyUI_Mira, mirabarukaso's utility pack, and the reasoning is the same as everywhere else in the pack: readability. A workflow that juggles a bunch of tunable numbers - LoRA strengths, blur amounts, mask intensities, whatever you've got - gets a lot cleaner when those values live in one labeled node instead of scattered across the canvas.

How it works

Eight float widgets, one list out. The node doesn't do arithmetic on them; it just packs them in order. Order matters, because the Float List Interpreter nodes on the receiving end pull values by index, so float_1 is index 0, float_2 index 1, and so on. Each field ranges from -10 to 10 in steps of 0.01, defaulting to 1, which covers the usual strength/multiplier range you'd want to distribute across a workflow.

Inputs and outputs that matter

• float_1 through float_8 - eight values, each -10 to 10, default 1. Set the ones you use; leave the rest at whatever's harmless for your graph.

The single output is float_list, a FLOAT_LIST. It's a bundle, not a scalar - don't wire it straight into something expecting one float. Send it to a Float List Interpreter (1 or 8 outputs) to get individual values back.

Installing it

ComfyUI Manager: search ComfyUI_Mira in the Custom Nodes Manager, Install, restart. Manual: cd ComfyUI/custom_nodes && git clone https://github.com/mirabarukaso/ComfyUI_Mira.git, then restart ComfyUI. If the pack errors on load, run pip install -r requirements.txt in its folder and restart. No models to download.

Common issues

The single predictable mistake: treating float_list like a plain float. It's a list; a node expecting one number won't read it correctly. Unpack it with an interpreter first.

Beyond that, mind the ranges. The fields cap at -10 to 10, so if you need a value outside that (a large multiplier, say), this isn't the node to source it from - it'll clamp you. And keep the index order in mind: float_1 sits at index 0, so when a Float List Interpreter's Start_At_Index reads "the first value," that's float_1. If an unpacked value comes out one slot off from what you set, it's almost always an index offset, not a bug. This is convenience plumbing; its only real job is to stay out of your way while keeping eight numbers tidy.