Skip to content
BytePlusBytePlusBytePlus ModelArk for ComfyUI

Generate and Enhance

Template
Generate and Enhance, in the ComfyUI template browser under BytePlus
Workflow file
Generate and Enhance.json
BytePlus nodes
Seedream 4.5 & 5.0, Image Quality Enhance, Seedance 2.5 Text to Video, vCube Video Enhance
Keys needed
ModelArk API key, AI MediaKit API key
Also uses
Compare Images, Save Image, Save Video (built into ComfyUI)

What it makes

Section titled “What it makes”

Generated media at a higher quality than the model outputs directly. The image group makes a product photo of a steaming ceramic coffee mug with Seedream 5.0 Pro and doubles its size with Image Quality Enhance. The video group, off by default, makes a 5-second clip of the same mug with Seedance and upscales it to 1080p with vCube Video Enhance.

The workflow

Section titled “The workflow”
Two groups: Seedream into Image Quality Enhance with Save and Compare Images, and below it bypassed Seedance into vCube Video Enhance
Image group on top (active), video group below (bypassed, purple).
  1. Seedream 4.5 & 5.0 makes the picture with seedream 5.0 pro at 1312x736 (16:9) from the prompt “A product photo of a ceramic coffee mug on a walnut table, morning window light…”.
  2. Image Quality Enhance takes it with tool_version standard and multiple 2.00, giving 2624x1472.
  3. Save Enhanced Image writes the result, and Compare Images shows the original output against the enhanced image in a slider (Nodes 2.0 only).
  4. Seedance 2.5 Text to Video (video group) makes the clip with the model Seedance 2.0 Fast at 720p, 16:9, 5 seconds, from a prompt about the mug and a clinking spoon.
  5. vCube Video Enhance (video group) upscales it with standard, scene aigc, style hd, resolution 1080p and the source frame rate. Save Enhanced Video keeps the video, and the source_frame and enhanced_frame outputs feed a second Compare Images.

Run it

Section titled “Run it”
  1. Add your ModelArk API key and your AI MediaKit API key in Settings → BytePlus if you have not yet (API keys). The template needs both, and they are separate keys.
  2. Log in to your Comfy.org account in ComfyUI. The enhance nodes upload the generated image or video to Comfy.org storage first, because AI MediaKit only takes links.
  3. Open the template browser and choose Generate and Enhance under BytePlus.
  4. Press Run. The image group runs and the enhanced picture appears in Save Enhanced Image.
  5. For the video, set the video group to Always and run again.

Settings worth changing

Section titled “Settings worth changing”
  • Seedream prompt and size_preset: describe your own subject. A larger preset gives a larger enhanced picture.
  • Enhance multiple: 2.00 is set. standard allows 1 to 8; professional and max go higher and add texture.
  • Seedance model: the video group uses Seedance 2.0 Fast. Pick another model for different speed or quality.
  • vCube resolution: 1080p is set. Change it, or fps, to change the output.
  • Video group: it is bypassed in the template. Set the group to Always to run it, or leave it off if you only need the image.

Example result

Section titled “Example result”
The enhanced photo of a steaming ceramic coffee mug on a walnut table in Save Enhanced Image
Example output from the image group.

Time and cost

Section titled “Time and cost”

ModelArk bills the Seedream image and the Seedance video, and AI MediaKit bills each enhancement. Groups that are bypassed do not run and do not bill. The seed controls are set to randomize, so every run makes new results and bills again. Set a control to fixed and ComfyUI reuses the cached result of an unchanged node without billing.

Section titled “Related”