Home
Creators
Search Recent Random
Posts
Search DMs Popular Hash Lookup Tags Random
Importer
Import FAQ
Register Favorites Keys Review DMs Logout
Community
Telegram ThePornDude
Home Creators Posts Import Register Favorites Logout

Detail Daemon! (Patreon)

Published:
2024-10-31 23:48:03
Edited:
2024-10-31 23:56:00
Imported:
2026-06
Tags:

Downloads

Content

Detail Daemon is an easy way to add details to Flux, SDXL or SD1.5 generations!

Video Guide: https://youtu.be/dG9HfgngsqM

GitHub Link

https://github.com/Jonseed/ComfyUI-Detail-Daemon

2 samplers, 1 stateless sigma multiply node and a graph

Use either the main Detail Demon sampler for the most detail, or the Multiply Sigmas node for an easy upgrade to any workflow :) 

Example image from GitHub:

Install nodes via ComfyUI Manager as normal:

 

Workflows

As usual, the workflows used in the video are attached.

Note the Pixart example is waaay down at the bottom, far below the Detail Daemon compare

Previous video featuring the Flux Unsampler - https://youtu.be/-qVDWE0A1ek

Files

Comments

Nibmeister

Awesome workflow. Thanks for posting.

LVisuals

Hi again <3, maybe this one is more accurate to replace magnific? Could add external image and upscale it?

simon gentry

wow! for the face! that skin change!

CJ

Hey there, new to the community. I've been following this specific walkthrough to add detail to some imagery I've developed in a 3rd party app. I'm having issues getting the detail daemon to adhere to my base image so I was hoping to try the fast unsampler method you showed. Is there a chance you could share that workflow?

GrzegorzBrzęczyszczykiewicz

Oh man, that is a solution that I was looking for for ages! That's absolutely awesome, thank you!

GrzegorzBrzęczyszczykiewicz

BTW, I’m wondering if replacing SamplerCustomAdvanced with ksampler or the Flux Sampler Parameters node (from Essentials) would make any significant difference. Do you know what the main differences are, apart from the presence or absence of a guider? Also, is there a notable difference between Empty Latent Image and EmptySD3LatentImage? I’ve seen the latter recommended a few times on YouTube, but I don’t think anyone has clearly explained why — at least not in simple terms. I'd really love to see a video that breaks down topics like this. Matteo from Latent Vision did an absolutely phenomenal job, but sadly, he doesn’t seem to be very active anymore.