Skip to main content
Video Upscale

Video Upscale · screenshot coming soon

​
What this node is meant to do

Video Upscale will upscale a COMFYTV_VIDEO frame by frame (like image Upscale, but for clips). scale reserves 2x / 4x. Output stays COMFYTV_VIDEO for Clip / Demux downstream.
⚠️ Current status: UI placeholder only. video_audio.py raises StageNotImplemented on Run — no fake URLs. See roadmap.md.

​
When to use it (once shipped)

  • Enlarge low-res I2V / T2V to 1080p+
  • Upscale plugin video (mesh2motion, etc.) inside ComfyTV
  • Upscale before subtitle erase (also pending)

​
Planned design

  • Same Stage + ▶ Run + snapshot model as other video stages.
  • Likely a workflows/ upscale graph or hybrid PyAV + model path (TBD).
  • Workaround now: Extract Frame → image Upscale → Create Video (native) → → ComfyTV Video.

​
Types

bridges.md

​
Parameters

​
video

Source COMFYTV_VIDEO (required when implemented).

​
scale

2x or 4x (when implemented).

​
Outputs (when implemented)

​
Step by step (today)

  1. Treat node as not available.
  2. For enlargement: Extract Frame + image Upscale per frame; watch roadmap.md.
  3. Do not expect Run to succeed.
This page covers one node only. For end-to-end workflows, multi-stage pipelines, type conversion, and design rationale, see the ComfyTV user guides on GitHub:

​
Repository and workflows

​
FAQ

Q: StageNotImplemented on Run?
A: Intentional placeholder — no fabricated outputs. Q: Upscale video today?
A: Extract Frame + Upscale per frame; or native upscale plugin + → ComfyTV Video. Q: 4x slower when shipped?
A: Likely yes; params have no effect now.
  • Upscale (image, ✅)
  • Extract Frame / → ComfyTV Video
  • Video Resize (geometric, ✅)