ComfyUI-Darkroom brings authentic film looks to AI images locally
ComfyUI-Darkroom is a custom node package for ComfyUI that provides professional color grading and film emulation capabilities. The suite includes 29 nodes covering film stocks, lens profiles, and comprehensive color adjustment tools—all processed locally without API costs.
Created by Jeremie Louvaert, a photographer and senior CG artist with years of animation production experience, this tool addresses a gap in the ComfyUI ecosystem. Many existing film emulation nodes apply simple color grades with stock names, but Darkroom uses actual technical data to achieve accurate results.
Built for accuracy and control
- 161 film stocks parsed from real Capture One curve data.
- Luminance-dependent grain that behaves like real film—coarser in shadows, finer in highlights.
- Physics-based halation modeled on light bouncing off the film base.
- 102 lens profiles with distortion and chromatic aberration data from real glass.
- Cinema print chain supporting Kodak 2383 and Fuji 3513 workflows.
Photographers, digital artists, and filmmakers working with AI-generated images can use these tools to achieve authentic film looks without relying on generic presets. The color grading nodes offer DaVinci Resolve-level control including lift/gamma/gain wheels, log grading, and hue-vs-hue adjustments.
Developer notes and future plans
Louvaert built Darkroom after becoming frustrated with existing options.
'Most ComfyUI film nodes are a vibe,'
he explained.
'A color grade with a stock name slapped on it. I wanted the real thing, so I built it.'
The package requires no GPU—processing happens through pure numpy and scipy computation in linear light space. All nodes support batch processing and include strength sliders for non-destructive blending. Louvaert continues adding profiles and asks users which stocks or lenses they use most, as that feedback will determine what he profiles next.
