vrgamegirl19 /comfyui-vrgamedevgirl

About comfyui-vrgamedevgirl

Adds custom nodes to ComfyUI, the node-based app for image and video models. The main one, AI Video Builder, works scene by scene with LTX 2.3 and MiniMax H3. You plan scenes from songs or SRT files, generate clips, fix faces, then stitch. It's source-available, and some tools need extra models you install yourself.

Best for: Video editors who want one node pack for planning, generating, repairing and assembling AI video projects like music videos and storyboarded shorts.

• Plans scenes from songs, audio, SRT files, lyrics or manual timing
• Builds image-to-video, text-to-video, ID-LoRA and First/Last Frame scenes
• Fixes faces and enhances clips, keeping the original resolution and frame count
• Creates captioned LoRA datasets and trains MiniMax H3, LTX 2.3 and Z-Image LoRAs
• Clones voices or turns text into speech with the VoxCPM2 node
• Adds LUT, grain, color match, sharpening and image comparison nodes

1. 1
   Install the node pack with ComfyUI Manager, or clone it into ComfyUI/custom_nodes, then restart ComfyUI.
2. 2
   Hard refresh the ComfyUI browser page so the latest JavaScript UI files load.
3. 3
   Add the VRGDG AI Video Builder UI node to open the Builder.
4. 4
   Create a project and add audio, SRT timing or scenes.
5. 5
   Use the Wizard or Storyboard Builder to plan the project.
6. 6
   Generate and approve scene images, render scene videos, then stitch the final video.