TrentNodes

AlignStylizedFrame

Align a stylized image to its original

Trent/Image · 3 inputs · 15 parameters · 5 outputs

AnimationDuplicateFrameProcessor

A ComfyUI node that processes animation frames to replace duplicate sequences with gray frames, making animation timing structure visible. Enhanced with multiple similarity metrics, frame padding insertion, and frame tracking.

Trent/Animation · 1 input · 19 parameters · 4 outputs

AnimationFrameRemover

A ComfyUI node that removes specified frames from an image batch. Used to remove padding frames inserted by AnimationDuplicateFrameProcessor.

Trent/Animation · 1 input · 2 parameters · 2 outputs

AudioLength

Calculates the length of audio in seconds.

Trent/Audio · 1 input · 0 parameters · 2 outputs

AudioToPhonemes

Extract phoneme timings from audio using Vosk speech recognition.

Trent/LipSync · 1 input · 2 parameters · 2 outputs

AutoStyleDataset

Outputs 35 prompt strings from auto_style_dataset_prompts.txt

Trent/Text · 0 inputs · 2 parameters

BatchSlowdown

Slow down video/image batches by intelligent frame duplication. Supports both integer and decimal multipliers with configurable distribution strategies for non-integer slowdowns. Modes: - multiplier: Direct slowdown factor (2.0 = 2x slower) - target_frames: Specify exact output frame count - fps_convert: Convert between frame rates (24fps -> 60fps) Distribution (for non-integer multipliers): - spread: Evenly distribute extra frames across the sequence - front_weighted: Extra frames concentrated at start - back_weighted: Extra frames concentrated at end

Trent/Video · 3 inputs · 8 parameters · 7 outputs

BevelEmboss

A ComfyUI node that applies bevel/emboss effects to images with enforced GPU support. Features extended parameter ranges and improved bevel algorithms.

Trent/Image · 2 inputs · 8 parameters · 1 output

BlackBarCinemaScope

Adds cinematic black bars to achieve widescreen aspect ratios.

Trent/Image · 1 input · 3 parameters · 2 outputs

BlackBarCrop

Detects letterbox/pillarbox black bars, crops them off, and

Trent/Image · 1 input · 4 parameters · 3 outputs

CherryPickFrames

Flexibly select frames from a batch using various modes: first N, last N, specific indices, or every Nth frame.

Trent/Image · 1 input · 4 parameters

ChopCuts

GPU-accelerated scene detection and video splitting node. Accurately detects cuts, fades, and transitions, then exports each scene as a separate MP4 file with a detailed report.

Trent/Video · 1 input · 10 parameters · 4 outputs

ChromaStripOverlay

Places a chroma key green strip on the side of a video

Trent/Image · 2 inputs · 5 parameters · 1 output

CompositionCowboy

Build print/cover layout prompts from curated, design-best-practice templates

Trent/Prompt · 8 outputs

ControlOverlayRodeo

Play an image batch with control maps overlaid on top.

Trent/Video · 5 inputs · 3 parameters · 1 output

CorridorKeyKeyer

Neural green screen keying using CorridorKey

Trent/Video · 3 inputs · 8 parameters · 3 outputs

CowboyMultiLineStringConcatenate

Joins up to ten strings into one string.

Trent/Text · 0 inputs · 1 parameter · 1 output

CowboyMultiLineStringConcatenateDynamic

Joins the connected inputs into one string.

Trent/Text · 0 inputs · 1 parameter · 1 output

CowboyNodeTitle

Connect any node output and get that node's title as a

Trent/Text · 1 output

CowboyWidgetRoundup

Connect nodes to the node_N inputs and type widget names into

Trent/Text · 0 inputs · 5 parameters · 1 output

CreatureLipSync

Complete lip sync pipeline: audio analysis, phoneme mapping,

Trent/LipSync · 4 inputs · 8 parameters · 3 outputs

CrossDissolveOverlap

Cross-dissolve two image batches with precise overlap control

Trent/Video · 1 input · 5 parameters · 3 outputs

CUN_NumberCounter

Compact description unavailable for this node.

Trent/Utilities · 0 inputs · 6 parameters · 3 outputs

CUN_TextFileLineLoader

Compact description unavailable for this node.

Trent/Utilities · 0 inputs · 4 parameters · 2 outputs

CustomFilenameGenerator

A ComfyUI node that generates custom filenames based on the format: PROJ_SEQ_SHOT_DEPARTMENT_WORKFLOW_PASS_INITIALS_VERSION

Trent/Utilities · 0 inputs · 10 parameters · 1 output

DrawGrayBBox

Fills each bbox region with neutral gray (0.5) and outputs a matching binary mask. Consumes the BOUNDING_BOX type from comfy-core's RT-DETR Detect / Draw BBoxes nodes.

Trent/Masks · 2 inputs · 1 parameter · 2 outputs

EasiestGreenScreen

One-click background removal and chroma key

Trent/Image · 2 inputs · 6 parameters · 2 outputs

FalKlingV2V

Calls FAL Kling O3 Pro video-to-video reference

Trent/API · 8 inputs · 7 parameters · 3 outputs

FileCollisionTestNode

A utility node to help test collision handling by creating dummy files

Trent/Utilities · 0 inputs · 2 parameters · 2 outputs

FileListNode

A node that lists files in a directory to help visualize collision handling results

Trent/Utilities · 0 inputs · 2 parameters · 1 output

FilenameExtractor

A ComfyUI custom node that extracts a portion of a filename. Searches for a pattern of 3 letters + 3 numbers (e.g., CRC040) and returns everything up to and including that pattern.

Trent/Utilities · 0 inputs · 0 parameters · 3 outputs

FirstValid

Returns the first truthy input value.

Trent/Flow

FrameRampBoogie

GPU-accelerated frame interpolation with

Trent/Video · 4 inputs · 6 parameters · 4 outputs

GetLatestImageFromFolder

Read the most recent image from a folder.

Trent/Video · 1 input · 5 parameters · 4 outputs

GrabFirstFrame

Returns the first frame from a batch of images.

Trent/Image · 1 input · 0 parameters · 1 output

GrayPaintTracker

Paint a gray region on the first frame of a video and have

Trent/Keyframes · 1 input · 13 parameters · 3 outputs

H3ChainAudio

Keep every extension's audio for the final VHS mux. Connect

Trent/MiniMax · 4 inputs · 0 parameters · 1 output

H3ChainStep

Run an exact number of extensions with a VHS Meta Batch Manager.

Trent/MiniMax · 1 input · 3 parameters · 5 outputs

ImageBatchAnalyzer

A ComfyUI node that analyzes a batch of images and generates comprehensive visual reports about their properties.

Trent/Image · 0 inputs · 2 parameters · 2 outputs

ImageColorWord

Output a single-word color name describing the image.

Trent/Image · 1 input · 2 parameters · 1 output

ImageFolderCowboy

Load images from a directory with natural sorting.

Trent/Image · 0 inputs · 10 parameters · 5 outputs

ImageSliceCowboy

Split an image / image batch into two halves along one axis.

Trent/Image · 1 input · 3 parameters · 4 outputs

ImageTextGrid

Creates a grid layout from a batch of images

Trent/Image · 10 inputs · 7 parameters · 1 output

IntOffsetClamp

Add offset to value and clamp to a minimum.

Trent/Utils · 1 input · 2 parameters · 1 output

JSONParamsExtractorNode

Compact description unavailable for this node.

Trent/Utilities · 0 inputs · 3 parameters · 1 output

JSONSummary

Compact description unavailable for this node.

Trent/Utilities · 0 inputs · 2 parameters · 1 output

JustMaskThosePads

Generate a mask highlighting only the padded regions

Trent/Image · 2 inputs · 1 parameter · 1 output

JustPadOrCropIt

Pad or crop an image to match a reference image's

Trent/Image · 2 inputs · 2 parameters · 2 outputs

LatentAlignedMask

Creates masks aligned to VAE latent space to prevent black spots and artifacts. Automatically calculates the correct blockiness based on VAE compression.

Trent/Masks · 2 inputs · 4 parameters · 3 outputs

LatentAlignedMaskAdvanced

Advanced version with character detection and smart expansion.

Trent/Masks · 2 inputs · 3 parameters · 2 outputs

LatentAlignedMaskSimple

Simplified version that doesn't require VAE input. Just specify compression factor directly (8 for Wan/SD models).

Trent/Masks · 1 input · 4 parameters · 2 outputs

LatentAlignedMaskWan

Optimized version specifically for Wan 2.1 VAE. Hardcoded for 8x spatial compression with sane defaults.

Trent/Masks · 1 input · 3 parameters · 1 output

LatestVideoFinalFrameNode

A ComfyUI node that finds the latest video file in a directory and extracts its final frame.

Trent/Video · 2 outputs

LatestVideoLastFramesNode

A ComfyUI node that finds the latest video file in a directory and extracts the last N frames from it.

Trent/Video · 3 outputs

LoRATestPromptGenerator

Generates 10 test prompts for validating LoRA models across

Trent/Testing · 0 inputs · 4 parameters · 11 outputs

MaskToBBoxMask

Per frame, finds the bounding box of the input mask (pixels above threshold) and

Trent/Masks · 2 inputs · 4 parameters · 3 outputs

MatAnyoneMatte

Temporally-consistent video matting using

Trent/Video · 3 inputs · 7 parameters · 2 outputs

MiniH3FrameAdjusterNode

Rounds a clip up to the next MiniMax H3-valid frame count ((n*17)+5:

Trent/Utilities · 2 inputs · 0 parameters · 5 outputs

MiniMaxH3RepackAVLatent

Rewrap a cross-run H3 AV latent (plain [video, audio]

Trent/MiniMax · 1 input · 0 parameters · 1 output

MiniMaxH3Resize

Resize an image to a MiniMax H3 output size (16:9 or

Trent/Image · 1 input · 4 parameters · 3 outputs

MouthShapeCompositor

Composite mouth shapes onto video frames for lip sync animation.

Trent/LipSync · 4 inputs · 5 parameters · 1 output

MouthShapeCompositorTracked

Composite mouth shapes onto moving characters using tracked

Trent/LipSync · 5 inputs · 9 parameters · 3 outputs

MouthShapeLoader

Load and organize mouth shape images for lip sync.

Trent/LipSync · 1 input · 1 parameter · 1 output

MouthShapePreview

Preview mouth shape sequence as text.

Trent/LipSync · 1 input · 1 parameter · 1 output

MultiBatchCombine

Concatenate multiple image batches into a single batch.

Trent/Image · 8 inputs · 4 parameters · 2 outputs

MultiLoadCowboy

Loads up to six images and resizes them all to one size.

Trent/Image · 3 inputs · 5 parameters · 11 outputs

name

Compact description unavailable for this node.

PatchSolAttention

Patches model self-attention with NVIDIA Sol-Attn (arXiv:2607.24027),

Trent/Optimization · 1 input · 4 parameters · 1 output

PhonemeToMouthShapes

Convert phoneme timing data to per-frame mouth shape sequence.

Trent/LipSync · 1 input · 5 parameters · 2 outputs

PointPreview

Preview and pick a point location on an image.

Trent/LipSync · 1 input · 3 parameters · 3 outputs

PointsToMasks

Convert point coordinates to tracking masks.

Trent/LipSync · 1 input · 3 parameters · 1 output

PointTracker

Track a point across video frames for mouth placement.

Trent/LipSync · 1 input · 7 parameters · 3 outputs

PSDBackgroundDetect

Detect the background layer range in a PSD by

Trent/PSD · 0 inputs · 5 parameters · 4 outputs

PSDLayerCompositor

Composite layers from a PSD Layer Splitter folder

Trent/PSD · 1 input · 16 parameters · 4 outputs

PSDLayerFlatten

Flatten a contiguous index range of PSD layers

Trent/PSD · 0 inputs · 5 parameters · 2 outputs

PSDLayerFlattenRGBA

Flatten a contiguous index range of PSD layers

Trent/PSD · 0 inputs · 5 parameters · 2 outputs

PSDLayerLoader

Load layer PNGs from a PSD Layer Splitter output

Trent/PSD · 0 inputs · 5 parameters · 5 outputs

PSDLayerNames

Open a .psd and list all its layer names. Pairs

Trent/PSD · 0 inputs · 3 parameters · 3 outputs

PSDLayerSaveAsPSD

Open an original .psd, replace one layer's

Trent/PSD · 1 input · 4 parameters · 2 outputs

PSDLayerSplitter

Open a .psd file and save every renderable layer

Trent/PSD · 0 inputs · 8 parameters · 5 outputs

RefFolderCowboy

Load up to six reference images from a folder onto six

Trent/Image · 1 input · 1 parameter

RefPreviewCowboy

Preview up to six images, silently skipping unconnected or

Trent/Image · 0 inputs · 0 parameters · 0 outputs

RemoveMouthBackground

Remove background from mouth shape images using BiRefNet

Trent/LipSync · 1 input · 8 parameters · 2 outputs

RoundUp

Always round up to the nearest whole number.

Trent/Utils · 1 input · 0 parameters · 2 outputs

SaveForNextTime

Save data into a named slot for a later queue run.

Trent/Utilities · 6 inputs · 3 parameters · 0 outputs

SaveTransparentVideo

Save Transparent Video

Trent/Video · 2 inputs · 4 parameters · 1 output

SmartFileTransferNode

A ComfyUI node that copies or moves files with intelligent name collision handling. If a file already exists, it automatically creates a unique name like file(1).ext, file(2).ext, etc.

Trent/Utilities · 0 inputs · 6 parameters · 3 outputs

StringListCowboy

Make a list of strings with optional prepend/append to each.

Trent/Text · 0 inputs · 3 parameters · 1 output

TakeFromLastTime

Read back data saved by Save for Next Time on an earlier

Trent/Utilities · 9 outputs

TextFileCowboy

Load one text file from a directory, keyed by filename

Trent/Text · 1 input · 4 parameters · 3 outputs

ThisThatOrTheOther

Parallel gate with 3 independent channels.

Trent/Flow

TranscribeLyrics

Transcribes audio with Whisper. Returns plain text, LRC-timed

Trent/Audio · 3 inputs · 6 parameters · 4 outputs

TrentCutDetective

Detects every shot boundary in a clip with OmniShotCut (2026

Trent/Video · 2 inputs · 12 parameters · 7 outputs

TrentEneasSegment

Track one object using text or points, or mask all objects in a category. White is foreground. Models download on first use.

Trent/Segmentation · 1 input · 7 parameters · 1 output

TrentGetEffingLastIndex

Reports the last frame index (count - 1) and the frame count for an IMAGE batch.

Trent/Utils · 1 input · 0 parameters · 2 outputs

TrentH3AudioSoundscaper

Hears the clip's audio with a local omni GGUF (Qwen3-Omni via

Trent/VLM · 1 input · 11 parameters · 5 outputs

TrentH3AutoPromptGenerator

Analyzes a video + identity reference image with a VLM and

Trent/VLM · 5 inputs · 24 parameters · 5 outputs

TrentH3CodexPromptor

Writes and refines MiniMax H3 prompts using the official MiniMax skill and your Codex ChatGPT login.

Trent/VLM · 5 outputs

TrentH3HermesPromptDirector

Uses Hermes Agent to research, draft, critique, and select a MiniMax

Trent/VLM · 0 inputs · 5 parameters

TrentH3LocalLLMStop

Stops the llama-server the H3 Skill Promptor spawned (or an

Trent/VLM · 1 output