WhatDreamsCost/WhatDreamsCost-ComfyUI

LTX Director and a variety of other custom ComfyUI nodes and workflows

WhatDreamsCost‑ComfyUI – Custom nodes for ComfyUI

What it is

A set of community‑made custom nodes that extend ComfyUI (a popular open‑source UI for diffusion models) with video‑ and audio‑focused workflow tools. The nodes let you load, trim, resize, and edit video/audio, apply prompt‑relay‑driven generation, and use IC‑LoRA techniques – all inside the visual node‑graph editor.

Core nodes

Node Main purpose Notable features
LTX Director 2.0 Full‑featured timeline editor Prompt‑Relay integration, first/middle/last‑frame support, IC‑LoRA tracks, audio in‑painting, retake mode, timeline save/load, UI overhaul, speed‑up optimisations (model caching, faster generations, Triton‑compatible attention).
Load Video UI Import and trim video clips Drag‑and‑drop, crop mode, aspect‑ratio presets, seconds↔frames toggle, filename output, reduced RAM usage.
Load Audio UI Import and trim audio files Drag‑and‑drop, duration setting, filename output, works with nodes 2.0 UI.
Multi Image Loader Gallery‑style image loader Inline gallery, batch or individual output, built‑in resize + LTXV preprocessing.
LTX Sequencer Shot‑sequence builder Syncs multiple sequencer nodes in real‑time, supports arbitrary numbers of middle frames, FFLF (first‑frame‑last‑frame) generation.
LTX Keyframer Alternate key‑frame editor Similar to Sequencer but kept for compatibility.
Speech Length Calculator Estimates video length from dialogue Real‑time length calculation based on quoted speech, updates without re‑running the workflow.

How to install

  1. Open the custom_nodes folder of your ComfyUI installation.
  2. Run:
    git clone https://github.com/WhatDreamscost/WhatDreamsCost-ComfyUI
    (or install via the ComfyUI Manager – make sure you fetch the nightly version if the latest release (v2.0.0+) isn’t listed yet).
  3. Update the companion nodes ComfyUI‑LTXVideo and ComfyUI‑KJNodes to their latest versions; the LTX Director node depends on them.

Typical workflow

  1. Load media – use Load Video UI or Load Audio UI to bring files into the graph.
  2. Arrange on timeline – drop the clips into LTX Director, add keyframes, prompts, or IC‑LoRA tracks.
  3. Add generation steps – connect Prompt‑Relay‑compatible nodes (e.g., text‑to‑image, image‑to‑video) to the timeline.
  4. Fine‑tune – use Speech Length Calculator to match video length to dialogue, enable Retake Mode to re‑generate specific segments.
  5. Export – render the final video/audio directly from the node graph.

Who might use it

  • Content creators who want AI‑augmented video editing without leaving a node‑based UI.
  • Researchers / hobbyists experimenting with text‑to‑video, audio in‑painting, or LoRA‑based style control.
  • ComfyUI power users looking for a richer video‑pipeline than the default loader nodes provide.

Maturity & support

  • Actively maintained (latest tag v2.0.4 as of the README). Recent updates focus on speed optimisations and UI polish.
  • Known issues: Retake mode is still experimental; large output videos can be RAM‑heavy due to ComfyUI limits.
  • The author provides a Ko‑fi link for donations and mentions that many nodes were prototyped with the help of Gemini (Google’s LLM).

Resources

  • YouTube tutorials – two linked videos demonstrate the LTX Director (v1.0 & v2.0) and a newer tutorial (https://www.youtube.com/watch?v=aXDIr8eNovI).
  • Example workflows – ready‑made .json pipelines are in the example_workflows folder of the repo.
  • Documentation – still being written; the README contains the most up‑to‑date feature list.

This summary is based solely on the information provided in the repository’s README.

Related

  • Project
    Lightricks/ComfyUI-LTXVideo

    A collection of custom ComfyUI nodes and workflows that extend the LTX-2 video generation model with advanced control, HDR output, AI dubbing, and generative upscaling.

  • Project
    vrgamegirl19/comfyui-vrgamedevgirl

    A collection of ComfyUI custom nodes featuring an AI Video Builder for scene-by-scene production, video enhancement, and LoRA training for AI video and music video creation.

  • Project
    yolain/ComfyUI-Easy-Use

    ComfyUI‑Easy‑Use is a genuine open‑source extension for the ComfyUI node‑based UI, providing dozens of convenience nodes (loaders, prompt tools, LoRA/ControlNet helpers, noise injection, image utilities, logic/loop nodes, etc.) that streamline building Stable Diffusion and other diffusion model workflows. It supports many models (SD‑1/2, SDXL, SD‑3, Flux, Kolors, Stable Cascade, etc.), offers UI theme tweaks, and includes documentation, video tutorials, and a workflow collection. Installation is a simple git clone into ComfyUI’s `custom_nodes` folder followed by running `install.bat`.

  • Project
    AlekPet/ComfyUI_Custom_Nodes_AlekPet

    A collection of custom nodes for ComfyUI that adds prompt translation, AI-driven content generation, and image manipulation tools.

  • Project
    huangserva/ComfyUI_MiniMaxH3_Director

    A collection of ComfyUI workflows for MiniMax H3 Director, enabling text-to-video, image-to-video, and advanced video-to-video editing with reference materials.