Loading
Have a ComfyUI workflow? Get a readable Workflow Doc in your browser.Make a workflow readable.Try it now ↗
This is a workflow that I've had very good success with and wanted to share. I started with the wonderful workflow found here: and adjusted it a bit. Props to this awesome workflow and if you are looking for a...
Runtime profile
Source description
This is a workflow that I've had very good success with and wanted to share. I started with the wonderful workflow found here: https://civitai.com/models/698637/comfy-ui-flux-dev-with-upscaler-2-loras-and-adetailer-super-clean and adjusted it a bit. Props to this awesome workflow and if you are looking for a complete txt2img solution, give that one a look.
This workflow is not an end to end workflow for txt2img. It is an img2img workflow. That's one of the biggest differences. Use this workflow once you already have an image ready.
Second, while there are prompts involved here (and optional LoRAs) there are two separate processes. The first is the optional face detailer process. This comes with its very own prompt and optional LoRAs. The second is the upscaler process which is also a detailer as you are upscaling the latent and not just the pixels.
AI-generated commentary
AI-generated explanation based on source and configuration details. Suggestions are clearly labeled.
ComfyUI Flux Detailer & Upscale — v1.0 is an img2img workflow for refining an existing image with optional face detailing and latent upscaling. It takes an existing image, separate prompts, and optional LoRAs for those stages, and can produce an image with enhanced detail at a4a8
It has two processes: an optional face-detailer stage and an upscaler stage that works on the latent rather than only the pixels.
The workflow loads an image, can apply face detailing, performs latent upscaling, and includes preview and save steps.
The face-detailing and upscaling stages use separate prompts and optional LoRAs while sharing the Flux.1 D base model.
Start with an image that is already available; this workflow is not an end-to-end txt2img workflow.
In a reported example, a 1920x1088 (2MP) Flux image was upscaled 2x to a 4K image.
The workflow uses Flux.1 D as its base model, is version v1.0, and is published by HackAfterDark.
The listed model files are: 4x-UltraSharp.pth, ae.safetensors, bbox/face_yolov8m.pt, clip_l.safetensors, flux1-dev.safetensors, fluxTextencoderT5XxlFp8_v10.safetensors, and sam_vit_b_01ec64.pth.
The listed node packs include: comfy-image-saver, comfyui_essentials, comfyui_ultimatesdupscale, comfyui-art-venture, and comfyui-custom-scripts.
The list also includes: comfyui-easy-use, comfyui-impact-pack, comfyui-impact-subpack, rgthree-comfy, and was-node-suite-comfyui.
Keep the face-detailing prompt and optional LoRAs separate from the upscaling prompt and optional LoRAs.
Suggestion · not verified
For Flux, consider starting with a 2MP image because lower-resolution images provide less to work with.
Suggestion · not verified
A reported RTX 3090 Ti run took 10–12 minutes for the upscale stage with a 2x upscale of a 1920x1088 (2MP) image; use this as a reference, not a runtime guarantee.
Suggestion · not verified
Memory needs for a 4x upscale are unknown; if you run out of memory, try two 2x stages. Before running, verify that 4x-UltraSharp.pth, fluxTextencoderT5XxlFp8_v10.safetensors, sam_vit_b_01ec64.pth, and was-node-suite-comfyui are present under those exact names.
Suggestion · not verified
Does this need to be edited?
Sign in to send an edit request.
Sources
1 sourceSource excerpts
2 excerptsSource context: 1102 downloads · Type Workflows · Base model Flux.1 D
Keeping the prompts and LoRAs separate is intentional and important (though both still use the same base model of course). You may want to adjust a face in an image with a LoRA designed for face details, jewelry, tattoos (face tattoo anyone?), etc. while you use a completely separate prompt and LoRA(s) for the upscaler process.
This is an incredibly powerful way to upscale and enhance the details of your images.
I can't overstate just how powerful this is. If you've never experimented with upscaling the latent space, please give this workflow a try as it may be an eye opener for you.
I'll share any updates I make in the future, but I've been using this workflow for a while now and am pretty happy with it. I don't want to add more nodes to it. I want to keep it simple and focused on details and upscaling.
Just a Heads Up!
Because it doesn't just upscale the pixels, this takes a while to run. On my RTX 3090 Ti, it took between 10-12 minutes for just the upscale stage (not using the face detailer stage) for a 2x upscale on a 1920x1088 (2MP) Flux generated image. This makes it a 4k image and it looks great, but it takes some time. Your needs may not involve an upscale to as high of a resolution, but I noticed for Flux, you're better off starting from a 2MP image. It will upscale lower resolution images, but there just isn't a lot to work with there. You could run it through a second time though.
In fact, you may need to run this in multiple stages if you run out of memory. In other words, if it doesn't work at a 4x upscale, run two 2x upscale jobs. I'm not entirely sure if you'll run out of memory or not, but it could take a really long time . I once accidentally set the input image to something that was already upscaled to 4k. It took me over an hour to realize my mistake, but it never crashed, I stopped the generation. So just be aware, this yields amazing results, but it can take some time.
Remaining hardware or download-size estimates are not available yet.
Requirements
Upscaler · 63.9 MB · PTH · Hugging Face · Kim2091/UltraSharp
VAE · Civitai · 1223425
Source context: 1074 downloads · Type Workflows · Base model Flux.1 D
Object Detector · 49.6 MB · Hugging Face · Bingsu/adetailer
Text encoder · 235 MB · SAFETENSORS · Hugging Face · diamkan/comfyui_oldmodel
Checkpoint · 22.2 GB · SAFETENSORS · fp32 · full · Civitai · 618692
fluxTextencoderT5XxlFp8_v10.safetensors
File unverifiedText encoder · Civitai · 704402 · SHA-256 7d330da48161
Segmentation · 358 MB · PTH · Hugging Face · comfyuistudio/sam_vit_b_01ec64.pth
comfy-image-saver
FoundNode pack · Registry
Node pack · Registry
comfyui_ultimatesdupscale
FoundNode pack · Registry
Node pack · Registry
comfyui-custom-scripts
FoundNode pack · Registry
comfyui-easy-use
FoundNode pack · Registry
comfyui-impact-pack
FoundNode pack · Registry
comfyui-impact-subpack
FoundNode pack · Registry
rgthree-comfy
FoundNode pack · Registry
was-node-suite-comfyui
PossibleNode pack · Registry
