Nodes/ComfyUI-ZML-Image/ZML_多文本输入_五
ComfyUI Node

ZML_多文本输入_五

Five prompt fragments in, one clean prompt out

By zml-w·Created about a year ago·Updated 2 months ago· 218
ZML_多文本输入_五
    • 文本
    格式化标点符号true
    分隔符,\n\n
    文本1
    文本2
    文本3
    文本4
    文本5

    The five-box middle child of the ZML multi-text-input family. Same idea as ZML_MultiTextInput3 - several multiline prompt boxes, a separator, punctuation cleanup - just with more slots, which is the right call once your prompts outgrow three fragments.

    Prompt engineering in practice isn't one paragraph; it's a stack of concerns you want to swap independently. Quality tags, subject, character details, style, artist - that's five. This node exists so each concern lives in its own box and you can swap one without rebuilding the whole string. It's the difference between editing a prompt and composing one.

    What's on it

    • 文本1 through 文本5 - multiline boxes, all optional. Empty ones are dropped from the join, so partial fills are fine.
    • 分隔符 (default ,\n\n) - the glue. The pack's house style is comma + two blank lines, and it reads well both in the box and in the encoded result.
    • 格式化标点符号 (default on) - runs the merged string through the pack's punctuation formatter: stray-comma cleanup, spacing normalization, NAI-weight-to-SD syntax conversion. Turn it off only if you're building something the formatter would step on.

    One output, 文本 (STRING), straight into a CLIP Text Encode.

    How it differs from the siblings

    • ZML_MultiTextInput3 - three boxes, merged string only. Grab it when you need fewer moving parts.
    • ZML_MultiTextInput5 - five boxes, merged string only. This one.
    • ZML_MultiTextInput5V2 - five single-line inputs, but five individual outputs plus a merged string and a list. That's the one for feeding distinct values to different nodes, or for the pack's text-image node.

    So if "five boxes, one prompt" is your shape, this is the one. If you need the pieces separately, skip to V2.

    Install

    It's one node in ComfyUI-ZML-Image:

    cd ComfyUI/custom_nodes
    git clone https://github.com/zml-w/ComfyUI-ZML-Image
    # restart ComfyUI
    

    or ComfyUI Manager → search "ComfyUI-ZML-Image". No dependencies beyond the pack baseline. Chinese-first UI - 格式化标点符号 is "format punctuation" - translation patch at https://github.com/zml-w/ZZZ_ZML_English_Patch for English users.

    These text nodes are the pack's most battle-tested territory (the pack literally started as a prompt-saving tool), so this one is reliable. The only real advice: make the five boxes a habit of structure - decide what each box means ("quality," "subject," "style") and keep the mapping consistent across your workflows. That's where the node stops being a convenience and starts being a system.

    Categoryimage/ZML_图像/文本

    Inputs (7)

    NameTypeDefaultDescription
    格式化标点符号COMBOtrue2 options: true, false
    分隔符STRING,\n\n
    文本1optSTRING
    文本2optSTRING
    文本3optSTRING
    文本4optSTRING
    文本5optSTRING

    Outputs (1)

    NameTypeDescription
    文本STRING