# mcmonkeyprojects/sd-dynamic-thresholding

Dynamic Thresholding (CFG Scale Fix) for Stable Diffusion (SwarmUI, ComfyUI, and Auto WebUI)

Repository: https://github.com/mcmonkeyprojects/sd-dynamic-thresholding
Canonical: https://ross.abutalabs.com/products/sd-dynamic-thresholding
Language: Python
License: MIT
License Family: permissive
Last push: 2025-03-14T09:33:32+00:00

## Health v2 (maintenance only)
Score: 36/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 11, release rhythm 35, longevity 93
- inputs: {"age_days": 1315, "days_push": 537, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1224, forks 107 (observed 2026-08-28T04:04:02.782618+00:00)

## What it is
A Stable Diffusion extension that enables using higher CFG scale values without color artifacts by clamping latents between sampling steps. It integrates with SwarmUI, ComfyUI, and AUTOMATIC1111 Stable Diffusion WebUI.

## Use cases
- use high CFG scale in stable diffusion without color burn
- fix saturated colors at high CFG values
- improve prompt adherence with stronger CFG
- enable dynamic thresholding in ComfyUI workflows
- add CFG fix to AUTOMATIC1111 webui
- generate images with extreme guidance scales

## When to choose
- you use SwarmUI, ComfyUI, or A1111 WebUI and want high CFG scales without artifacts
- your images get oversaturated or color-shifted at high CFG values
- you want better prompt following via stronger guidance

## When to avoid
- you use a diffusion UI other than SwarmUI, ComfyUI, or A1111
- you only generate at low CFG scales where the fix is unnecessary
- you need a standalone image generator rather than an extension

## Facets
- artifact type: plugin
- maturity: active
- function: image-processing, machine-learning
- domain: image-processing, artificial-intelligence
- platform: python, cross-platform
- tags: stable-diffusion, cfg-scale, webui-extension, comfyui, swarmui, dynamic-thresholding

## Member repositories
- mcmonkeyprojects/sd-dynamic-thresholding (main) score 36

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:02.782618+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 2026-08-30T06:15:28.774515+00:00, confidence not recorded.
  - readme: https://github.com/mcmonkeyprojects/sd-dynamic-thresholding (fetched 2026-08-28T04:04:02.782618+00:00, sha 484a1ad39c32)
- Data as of 2026-08-30T08:39:29.467469+00:00.