Nodes/comfyui_LLM_party/Parameter Dictionary Function
ComfyUI Node

Parameter Dictionary Function

Build one key:value pair for this pack's dict combiners

By heshengtao·Created 2 years ago·Updated 6 days ago· 2,321
Parameter Dictionary Function
  • value
  • parameter
key

The building block underneath this pack's Parameter Dictionary Combine nodes. On its own it does one small thing: wrap a key and a value into a single-entry dictionary, {key: value}. It only becomes interesting once you notice what value actually accepts.

How it works

key is a plain string, the name of the field. value is typed * - the ComfyUI wildcard - meaning it accepts anything: a string, a number, an image, a conditioning, whatever's wired in. That's the actual point of this node: a DICT's value can hold any type, so this is how you get a mixed-type field into a parameter object without ComfyUI's normal type-checking getting in the way.

The inputs and outputs that matter

  • key (default empty) - the field name. Required, but the empty default means it's easy to skip setting it - don't.
  • value (*) - anything you wire in. Required.

One output: parameter (DICT) - a single-key dictionary, ready to feed into parameter_combine or parameter_combine_plus alongside others built the same way.

Where it fits

Use one of these per field you need in a structured parameter object - a name field from a string, a count field from an int, an image field from actual image data - then merge them together with one of this pack's Combine nodes. This node's category, "APP link," is a hint at the intended use: assembling the arguments for calling out to an external app or action, one named field at a time.

How to install it

Ships with the full pack:

  • ComfyUI Manager: search "comfyui_LLM_party", install, restart.
  • Manual: cd ComfyUI/custom_nodes && git clone https://github.com/heshengtao/comfyui_LLM_party, then pip install -r requirements.txt from inside the pack folder using ComfyUI's own Python, then restart.

Pure data wrapping, no external dependency of its own.

Common issues & troubleshooting

key was left at its blank default. You end up with a dict keyed by an empty string, which is technically valid but useless once it's merged with others or handed to whatever's supposed to read it by name. Always set a real key.

Downstream doesn't recognize the field. This node doesn't validate or suggest key names for you - if whatever consumes the combined dict (an external API, another node's expected schema) needs a specific field name, match it exactly, including case.

value's type looks wrong once combined. Since value accepts anything, it's easy to accidentally wire in the wrong upstream output - double-check what type is actually flowing into it if the downstream consumer complains about a field's shape rather than its presence.

Category大模型派对(llm_party)/APP链接(app link)

Inputs (2)

NameTypeDefaultDescription
keySTRING
value*

Outputs (1)

NameTypeDescription
parameterDICT