Int to List Converter
Bundle a pile of ints into one list wire
- int_list
Some workflows need a handful of numbers to travel as one unit - a schedule of steps, a batch of seeds, a set of widths for a multi-pass pipeline. ComfyUI doesn't make that natural; you'd normally be wiring five separate int widgets into five separate nodes. Int to List Converter collapses that: up to twenty int inputs in, one list out.
How it works
You get int_input_1 (required) plus int_input_2 through int_input_20 (optional), and a length control that says how many of them to actually pack. Set length to 3 and only the first three inputs are collected into the list; the rest are ignored. Unconnected inputs within your chosen length silently default to 0 rather than erroring, which is either convenient or a subtle trap depending on how careful you are.
The output is a real Python list flagged with OUTPUT_IS_LIST, which is what makes it different from just concatenating ints - downstream nodes that accept a list can iterate it directly.
Inputs
- int_input_1 - the one required input.
- int_input_2 … int_input_20 - the optional expansion. The whole point of this node is that you can connect as many as you need and never touch a config.
- length - how many inputs get collected, 1 to 20.
Output
int_list - the packed list, typed as a wildcard (*) so any list-consuming node will take it.
Install
Standard pack install - ComfyUI Manager (search "ComfyUI-AutoFlow") or:
cd ComfyUI/custom_nodes
git clone https://github.com/ZXL-Xinram/ComfyUI-AutoFlow
Restart after. No dependencies beyond torch (already present).
Where people get burned
- The
lengthdefault is 1. If you connect five inputs and forgetlength, you get a one-element list. This is the classic footgun - always setlengthto match the inputs you actually wired. - The 0-fill for gaps.
length5 with only inputs 1–3 connected yields[a, b, c, 0, 0], not an error. If you see surprising zeros downstream, that's why. - It only converts ints. Floats, strings, booleans? Wrong node. Keep it to integers.
The realistic use is feeding the pack's own sibling nodes - List to Int Extractor and List Length - to round-trip values, or handing a batch of ints to a loop-style node that consumes lists. It's a plumbing utility: boring, and exactly what you want when you need it.
Inputs (21)
| Name | Type | Default | Description |
|---|---|---|---|
| int_input_1 | INT | 0-999999–999999 | — |
| length | INT | 11–20 | — |
| int_input_2opt | INT | 0-999999–999999 | — |
| int_input_3opt | INT | 0-999999–999999 | — |
| int_input_4opt | INT | 0-999999–999999 | — |
| int_input_5opt | INT | 0-999999–999999 | — |
| int_input_6opt | INT | 0-999999–999999 | — |
| int_input_7opt | INT | 0-999999–999999 | — |
| int_input_8opt | INT | 0-999999–999999 | — |
| int_input_9opt | INT | 0-999999–999999 | — |
| int_input_10opt | INT | 0-999999–999999 | — |
| int_input_11opt | INT | 0-999999–999999 | — |
| int_input_12opt | INT | 0-999999–999999 | — |
| int_input_13opt | INT | 0-999999–999999 | — |
| int_input_14opt | INT | 0-999999–999999 | — |
| int_input_15opt | INT | 0-999999–999999 | — |
| int_input_16opt | INT | 0-999999–999999 | — |
| int_input_17opt | INT | 0-999999–999999 | — |
| int_input_18opt | INT | 0-999999–999999 | — |
| int_input_19opt | INT | 0-999999–999999 | — |
| int_input_20opt | INT | 0-999999–999999 | — |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| int_list | * | — |