ComfyUI Node

D2 XY Seed2

Need 20 seeds, 10 incrementing, or 8 random? Count them out instead of typing

By da2el-ai·Created 2 years ago·Updated about a month ago· 66
D2 XY Seed2
    • x / y_list
    • LIST
    • BATCH
    initial_number-1
    count1
    moderandomize

    The other seed node in the D2 pack, and the one you want when the list is long or follows a pattern. Where D2 XY Seed makes you type every seed yourself, D2 XY Seed2 generates the whole list for you: give it a starting number, a count, and a mode, and it produces count seed values ready for an XY plot. Great for "I want to run this grid across 12 random seeds" or "I want a controlled sweep that increments by one."

    The inputs that matter

    • initial_number - the reference value (default -1). For fixed, increment, and decrement modes this is the base. If you pass -1 in any mode that uses it, it first resolves to a random seed, then applies the mode from there.
    • count - how many seeds to generate (default 1). With count = 0 you get an empty list, which is almost certainly a mistake.
    • mode - fixed (all the same value), increment (base, base+1, base+2…), decrement (base, base−1…), or randomize (a fresh random seed for each, ignoring initial_number). Default randomize.

    Outputs - same trio as D2 XY Seed: x / y_list (newline-joined, for the plot), LIST (array), and BATCH (list output for feeding multiple seeds at once). Set the plot's x_type/y_type to INT so the values stay numbers.

    How to use it

    Random batch: initial_number = -1, count = 12, mode = randomize → twelve distinct random seeds. Wire x / y_list into the plot's x_list. That's the whole thing.

    The increment/decrement modes are the ones that look like they'd be niche and then save you: a 1.5 sweep across initial_number = 12345, count = 5, mode = increment gives you 12345…12349 - useful when you want reproducibility and a small controlled spread. Just remember the initial -1 resolution happens first, so increment from -1 is "a random seed, then +1 steps from it" - deliberate, but surprising if you expected counting from zero.

    Installing

    Part of the D2-nodes-ComfyUI pack - install the pack, not the node.

    • ComfyUI Manager → search "D2-nodes-ComfyUI" → Install → restart.
    • Or cd ComfyUI/custom_nodes && git clone https://github.com/da2el-ai/D2-nodes-ComfyUI and restart.

    Dependencies: piexif, charset-normalizer - nothing heavy, no model downloads. Pack gotcha: v32.0.0+ requires the newer ComfyUI V3 schema; older ComfyUI → pre-32.0.0 release.

    Where people get burned

    Three things. randomize deliberately ignores initial_number, so don't expect a seeded random sequence from it - you get new randomness every run (fine for "give me options," not for "reproduce this grid"). count of 0 produces nothing, and there's no guard, so check it if the plot comes back empty. And the BATCH output is a list-typed output meant for nodes that accept multiple values at once - it doesn't do anything special for the plot, which wants x / y_list.

    CategoryD2/XY Plot

    Inputs (3)

    NameTypeDefaultDescription
    initial_numberINT-1-1–18446744073709550000
    countINT10–18446744073709550000
    modeCOMBOrandomize4 options: fixed, increment, decrement, randomize

    Outputs (3)

    NameTypeDescription
    x / y_listSTRING
    LISTLIST
    BATCH*