SeedraAI / launch
The template all of this runs on.
A ready ComfyUI image on Vast.ai with the nodes, dependencies and folders already in place. It boots in a couple of minutes and you pay only for the hours the card runs.
Open the template on Vast.aiUsing this partner link supports SEEDRAAI: Vast.ai shares a commission with us, while your price stays the same.
Current Vast.ai offer examples: RTX 5090 32 GB around $0.59/hr; RTX PRO 6000 96 GB around $1.18/hr.
Live price and availability vary by host and may change at any time.
Why this one
- The model installer only runs here
The command pulls models into the exact paths the image provides. On a bare machine it simply will not complete.
- Nodes and dependencies are already installed
No red nodes to chase and no version roulette: the workflow opens and runs.
- A card for the job, not forever
Photo workflows start at 8 GB VRAM; video workflows require 32 GB or more. Rent the card for the hours you need, then stop the instance.
The template is public. It is not tied to a purchase: you can boot it and run your own workflows without buying anything from us.
Running a pack
- Open the template and start an instance with the card you need.
- Run the model install command that ships with the pack.
- Load the workflow .json in ComfyUI and run a short test.