From 2271463fbb734ef889043fbaef7137acc1472627 Mon Sep 17 00:00:00 2001 From: Ramon Guthrie <156688666+RamonGuthrie@users.noreply.github.com> Date: Mon, 6 Jul 2026 11:37:59 +0100 Subject: [PATCH] Update README.md --- README.md | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/README.md b/README.md index 13e4177..951b2fe 100644 --- a/README.md +++ b/README.md @@ -4,7 +4,8 @@ ## Advanced Seed Diversity Enhancement for ComfyUI ✨ -Some generative models (like Krea 2, Z-Image Turbo and Qwen-Image) suffer from **limited seed variance & diversity bias** , changing the seed (especially in single-step or Turbo workflows) often produces nearly identical compositions or subtle variations. Conversely, highly sensitive multi-encoder models (like Flux or Wan 2.2) can easily collapse or lose prompt alignment when raw noise is injected. The **RBG Smart Seed Variance** node solves this by injecting targeted, mathematically structured noise directly into text embeddings during early generation steps. It creates meaningful diversity, styling variation, and composition shifting while preserving your core prompt intention. +Some generative models (like Krea 2, Z-Image Turbo and Qwen-Image) suffer from **limited seed variance & diversity bias** — changing the seed (especially in single-step or Turbo workflows) often produces nearly identical compositions or subtle variations. Conversely, highly sensitive multi-encoder models (like Flux or Wan 2.2) can easily collapse or lose prompt alignment when raw noise is injected. +The **RBG Smart Seed Variance** node solves this by injecting targeted, mathematically structured noise directly into text embeddings during early generation steps. It creates meaningful diversity, styling variation, and composition shifting while preserving your core prompt intention. --- Screenshot 2025-12-09 140138