live-coding-academy_ ⌕ search
home/ atoms/ comfyui-prompt-emphasis-weighting

ComfyUI weights a prompt term with (term:weight) syntax to strengthen or weaken it

Fact L2First instrument K · AI & real-time generative AV

In ComfyUI text prompts, wrapping a word or phrase in parentheses with a colon and number, like (good code:1.2) or (bad code:0.8), scales how strongly the CLIP text encoder emphasizes that term. Values above 1.0 push the diffusion toward that concept; values below 1.0 suppress it. Bare parentheses () default to 1.1. Literal parentheses in a prompt must be escaped as ( or ). This gives fine per-term control over image content without changing the whole prompt, and is a core prompt-crafting skill for anyone driving diffusion pipelines live.

Examples

Prompt “a (neon:1.4) city at (night:0.7)” boosts neon and softens the night cue. “(good code:1.2)” raises weight to 1.2; “(bad code:0.8)” lowers it to 0.8; a bare “(word)” applies the 1.1 default.

Assessment

Given the prompt “a (foggy:1.3) forest”, state what the 1.3 does and predict the effect of lowering it to 0.6. Then write a prompt that strongly emphasizes one term and weakly de-emphasizes another using the correct syntax.

Prerequisites — understand these first

Don't confuse with

Sources

“You can use () to change emphasis of a word or phrase like: (good code:1.2) or (bad code:0.8). The default emphasis for () is 1.1.”
corpus · comfyui-modular-node-graph-engine-for-diffusion-pipelines · chunk 13

Taught in modules