ComfyUI-tbox
Nodes:ImageLoader, ImageSaver, ImagesSaver, ImageResize, ImageSize, GFPGANNode, MaskAddNode, Video Load, ...
Nodes (21)
OpenPose, but for animals
Process long videos in chunks instead of all at once
The classic edge detector, and it needs no model at all
Keep a batch of images inside a size box
Per-pixel body maps for pose and animation control
Tbox's built-in pose extractor for ControlNet
The old Restore Faces GAN, and when it still beats ADetailer
A file loader that hands Qwen-VL a path, not just pixels
Image Resize (ComfyUI Essentials)
Tbox's single-path image export, and how it differs from its plural sibling
The tiny plumbing node that reads width, height, and batch count
Tbox's batch export node, and how not to grab its lookalike
Natural outlines for photo-based ControlNet
The no-download, all-purpose line extractor
Union up to seven masks into one
The old-reliable depth map, still bundled in tbox
The OOM-fighting utility node people wire behind every KSampler
Pick a video from your input folder for Qwen3-VL
Tbox's video export node, and why nvenc isn't always the right pick
Stamping a logo onto your generations (the opposite of watermark removal)