comfy-videogen
comfy-videogen
Use this skill for video generation and video upscaling through the
comfy-videogen CLI. Local LTX 2.3, WAN 2.2, and MiniMax H3 modes use model files under
/mnt/models/comfyui. RTX upscaling uses nvidia-vfx, not model files.
SeedVR2 upscaling uses its pinned upstream CLI and auto-downloads its own model
files. Remote Seedance 2.0 modes use ComfyUI API Nodes vendored by
comfy-diffusion and require COMFY_ORG_API_KEY.
If a built-in local video profile is missing files, use comfy-model-downloader
for the requested local videogen.<mode> capability before running inference.
Do not use the downloader for Seedance 2.0, RTX upscaling, or SeedVR2
upscaling; seedance2-api, rtx-vsr, and seedvr2 are not managed by
comfy-models download.
The CLI is quiet by default and prints only final JSON. Use --verbose only when
debugging ComfyUI runtime output, warnings, or progress bars.