🧠 Ollama & LM Studio Prompt From Idea

A ComfyUI node in LLM Prompts with 14 inputs and 3 outputs.

By NeonLightning·Created about a year ago·Updated about a year ago· 2

◄model▾►

◄ideafuturistic cyberpunk city►

◄negative►

◄max_tokens75►

◄min_tokens50►

◄max_attempts30►

◄regen_on_each_usetrue►

◄just_use_ideafalse►

◄exclude_commafalse►

◄randomize_seedtrue►

◄llm_seed207824427►

◄randomize_temptrue►

◄llm_temp0.59►

◄keepllmfalse►

CategoryLLM Prompts

Inputs (14)

Name	Type	Default	Description
model	COMBO		Select the LLM model (Ollama or LM Studio) to generate prompts with.
idea	STRING	futuristic cyberpunk city	Enter the core concept or theme for your prompt. You can have separated ideas if you have a hard return. Only use up to 3 lines though, to a maximum of 231 tokens.
negative	STRING		Words or themes to exclude from the prompt (used by Stable Diffusion, not LLM).
max_tokens	INT	7510–1024	Maximum token length for the generated prompt.
min_tokens	INT	5010–1024	Minimum token length for the generated prompt.
max_attempts	INT	301–200	Number of attempts to generate a prompt fitting token limits.
regen_on_each_use	BOOLEAN	true	Force regeneration on each node execution (doesn't matter if just_use_idea is on).
just_use_idea	BOOLEAN	false	Skip Generating and just use idea as prompt.
exclude_comma	BOOLEAN	false	Disables commas and sentence removal suggesting.
randomize_seed	BOOLEAN	true	Use a random seed on each generation.
llm_seed	INT	2078244270–999999999	Fixed seed (only used if randomize_seed is off).
randomize_temp	BOOLEAN	true	Use a random temperature on each generation.
llm_temp	FLOAT	0.590.1–1	Fixed temperature (only used if randomize_temp is off).
keepllm	BOOLEAN	false	Keep LLM model in memory.