Skip to content
WorkflowListedExternal
Open source

Ultra Upscaler

Personal upscaler. With the default settings, 720p images (realistic) with a good amount of detail work best. For anime and other content, lower settings may suffice. However, you are free to change the parameters.

Sourcecivitai.com/models/2020911
TaskMultiCapabilitiesImage edit · Image to image · UpscalingFamilyStable DiffusionEstimated VRAMNot yet characterizedBgrade

Runtime profile

Access
External workflow
Status
Listed
Source
Civitai
Updated
Oct 24, 2026

Source description

Personal upscaler. With the default settings, 720p images (realistic) with a good amount of detail work best. For anime and other content, lower settings may suffice. However, you are free to change the parameters.

Adjust the First Stage Denoise and Second Stage Denoise nodes to control how much the model re-imagines details. Higher values add more detail but can alter the image more. A good range is 0.4-0.7.

Used resources: https://civitai.com/models/277058/epicrealism-xl

https://huggingface.co/xinsir/controlnet-tile-sdxl-1.0

https://huggingface.co/ByteDance/SDXL-Lightning/blob/main/sdxl_lightning_8step_lora.safetensors

What you need

Hardware and download-size estimates are not available yet.

Requirements

Models

4

Node packs

5

AI-generated commentary

AI-generated explanation based on source and configuration details. Suggestions are clearly labeled.

Purpose

  • Ultra Upscaler is a personal image-upscaling workflow: it takes an image and produces an upscaled image, with realistic 720p images that have good detail described as the best fit with default settings.

Flow

  • The First Stage Denoise and Second Stage Denoise nodes control how much detail the model re-imagines.

Inputs and outputs

  • For anime and other content, lower settings may suffice.

Setup

  • The base model is SDXL 1.0.

Dependencies

  • Listed model files: bbox/face_yolov8m.pt, controlnet-tile-sdxl-1.0.safetensors, epicrealismXL_vxviiCrystalclear.safetensors, and sdxl_lightning_8step_lora.safetensors.

  • Named ComfyUI node packs: comfyui_essentials, comfyui-impact-pack, comfyui-impact-subpack, comfyui-kjnodes, and comfyui-wd14-tagger.

Configuration

  • The described denoise range is 0.4–0.7; higher values add detail but can alter the image more.

Cautions

  • Before running, check that your setup includes the four listed model files and five named ComfyUI node packs.

    Suggestion · not verified

Does this need to be edited?

Sign in to send an edit request.

Sources

1 source
CivitaiPrimary
civitai.com/models/2020911

Verified Jul 19

Source excerpts

1 excerpt
NoteJul 19civitai.com/models/2020911

Source context: 550 downloads · Type Workflows · Base model SDXL 1.0