ComfyUI-Wan-VACE-Prep

ConstrainedInt

An integer primitive with per-node min, max, and step constraints. The value snaps to the nearest step inside the bounds; the step grid is counted from min when set, otherwise from zero.

0 inputs · 4 parameters

Duration

Converts a requested duration in seconds to a MiniMax H3-compatible frame count, rounded up to the nearest 17k+5 at 24 fps, and returns the revised duration.

0 inputs · 1 parameter

FrameNumberOverlay

Burns a frame number label into every frame of an IMAGE batch as a text overlay. Useful for identifying frames in long sequences or debugging video workflows. Parameters: | Parameter | Default | Description | |-|-|-| | images | | Input image batch (IMAGE) | | fontsize | 32 | Size of the overlay text (8–256) | | startindex | 0 | Starting frame number for the first frame in the batch | | position | top-left | Placement: top-left, top-right, bottom-left, or bottom-right | | padding | 10 | Distance from edge in pixels | | fontcolor | white | Text color (any CSS color name or hex) | | prefix | "" | Optional text before the frame number | | suffix | "" | Optional text after the frame number | Outputs: | Output | Description | |-|-| | images | Image batch with frame numbers burned in |

1 input · 7 parameters

LoadVideosFromFolderSimple

Loads all videos from a folder, concatenated into a single image batch with their audio tracks combined. Optionally connect a VideoHelperSuite Meta Batch Manager node to process large collections in RAM-safe chunks. If you are joining a large number of video files and running out of system memory as they concatenate, this is the solution. From the VHS Meta Batch Manager node documentation: > The Meta Batch Manager allows for extremely long input videos to be processed when all other methods for fitting the content in RAM fail. It does not affect VRAM usage. It must be connected to at least one Input (a Load Video or Load Images) AND at least one Video Combine. See the VHS Meta Batch Manager node documentation for more information. Meta Batch Manager rule of thumb: set framesperbatch to roughly 10× your available RAM (not VRAM) in GB. So 32 GB -> 320 frames, 64 GB -> 640 frames, 128 GB -> 1280 frames. - Formats: webm, mp4, mkv, gif, mov - All videos must have identical resolution - No external dependencies Parameters: | Parameter | Default | Description | |-|-|-| | folderpath | | Full pathname of the directory holding input videos | | debug | false | Log video details and progress to the console | | metabatch | (optional) | Connect to VideoHelperSuite Meta Batch Manager to load videos in batches | Outputs: | Output | Description | |-|-| | images | Concatenated image batch ready for video creation | | audio | Combined audio track from all loaded videos. Disabled (empty) if any video lacks audio. | This node may be fragile under Nodes 2.0 / V3. The optional metabatch input us…

0 inputs · 2 parameters

VACEOutpaint

Provides an interactive outpaint layout that generates a VACE control video and binary outpaint mask from source frames; portions of the output window extending beyond the source are marked for outpainting.

1 input · 5 parameters

WanFMLF2V

Compact description unavailable for this node.

WanVACEBatchContext

Establishes iteration context for batch video-join processing. It selects an adjacent video pair or, for the loop-closing iteration, the last and first videos; it also produces work paths and iteration flags.

0 inputs · 6 parameters

WanVACEExtend

Generates a VACE control video and mask to extend a video from a selected frame position using preceding context frames.

1 input · 3 parameters

WanVACEFirstMiddleLast

Builds a VACE control video and mask from optional first, middle, and last frame batches. Known frames are placed at their positions with mask=0; remaining frames are gray placeholders with mask=1 for generation.

3 inputs · 4 parameters

WanVACEInpaint

Prepares a video for VACE inpainting by replacing masked regions with gray placeholders while preserving the rest of the video.

2 inputs · 0 parameters

WanVACEPrep

A small collection of ComfyUI nodes for common video tasks. Primarily designed for Wan VACE, with LTX-2 outpainting support. Note: As of v1.1.0, all code has been migrated to ComfyUI's Nodes 2.0 (V3) API. Nodes are still 100% compatible with the legacy ComfyUI renderer. There should be no noticeable changes in behavior, but please report any problems you encounter. v1.0.25 is the last legacy (pre-V3) release. Install via ComfyUI Manager: Search for "Wan VACE Prep" Or clone this repository: bash cd /path/to/comfyui/customnodes git clone https://github.com/stuttlepress/ComfyUI-Wan-VACE-Prep - Video Outpaint (formerly VACE Outpaint) - VACE Join - VACE Join (Batch) - VACE Batch Context - VACE Extend - Load Videos From Folder (Simple) - MiniMax H3 Duration - Int (Constrained) Experimental: - VACE Inpaint - Frame Number Overlay - Wan First/Middle/Last Frame to Video - VACE First/Middle/Last Prepares a video for outpainting using an interactive canvas widget. Position and size an output window over your source frames. Regions outside the source become the outpaint area. A model selector targets Wan VACE, LTX-2, or MiniMax H3: each sets its own quantisation grid and pad color. Renamed from VACE Outpaint. Parameters: | Parameter | Default | Description | |-|-|-| | images | | Source video frames | Canvas controls: | Control | Description | |-|-| | output resolution | Width and height of the generated output. Leave at 0 to match the crop box size. | | model | Consumer model. Sets the quantisation grid and the pad color together: wan = 16px grid, gray (0.5); ltx = 32px grid, black (0.…

2 inputs · 3 parameters

WanVACEPrepBatch

Performs batch-aware VACE preparation for video pairs, handling first- and last-iteration edge cases when producing transition controls and preserved start/end segments.

2 inputs · 6 parameters