# kijai/ComfyUI-SUPIR

SUPIR upscaling wrapper for ComfyUI

Repository: https://github.com/kijai/ComfyUI-SUPIR
Canonical: https://ross.abutalabs.com/products/comfyui-supir
Language: Python
License: NOASSERTION
License Family: other
Last push: 2026-04-29T15:30:56+00:00

## Health v2 (maintenance only)
Score: 61/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 79, release rhythm 35, longevity 65
- inputs: {"age_days": 917, "days_push": 126, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2308, forks 125 (observed 2026-08-28T04:06:35.879864+00:00)

## What it is
A ComfyUI custom node wrapper for the SUPIR image upscaler, an SDXL img2img pipeline with a custom ControlNet. It offers more efficient model loading, lower memory usage, wider hardware support, and more sampler options than the original implementation.

## Use cases
- upscale low-resolution images with SUPIR in ComfyUI workflows
- restore and enhance old or blurry photos using diffusion-based upscaling
- build image restoration pipelines with SDXL img2img
- skip or replace SUPIR's first-stage denoising with custom preprocessing nodes
- run fast Lightning-model upscaling for quick experimentation

## When to choose
- you already use ComfyUI and want SUPIR upscaling as nodes
- you need more sampler options and lower memory usage than the original SUPIR demo
- you want to combine SUPIR with LoRAs and other ComfyUI nodes

## When to avoid
- you want actively developed SUPIR support - it's now in ComfyUI core and this wrapper only gets breaking bugfixes
- you don't use ComfyUI
- you have limited VRAM and high-resolution inputs, since memory needs scale with input resolution

## Facets
- artifact type: plugin
- maturity: maintenance
- function: image-processing, machine-learning, plugin-system
- domain: image-processing, artificial-intelligence, media
- platform: python, cross-platform
- tags: comfyui, supir, upscaling, sdxl, custom-nodes, img2img, gpu

## Member repositories
- kijai/ComfyUI-SUPIR (main) score 61

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:35.879864+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:39:37.585064+00:00, confidence not recorded.
  - readme: https://github.com/kijai/ComfyUI-SUPIR (fetched 2026-08-28T04:06:35.879864+00:00, sha 01daa486fd94)
- Data as of 2026-08-30T08:39:29.467469+00:00.