1. Hourly Charges for Running ComfyUI Instances
When you launch and run a ComfyUI instance, charges are calculated based on the GPU size you select.
How It Works:
Charges are applied on an hourly rate.
Billing is done in 30-second increments, ensuring precise cost calculations.
The rate depends on the GPU size chosen when starting the instance.
How to Manage Costs:
Charges apply only while the instance is active and running.
To stop charges, turn off your instance from the My Instances page.
Charges stop immediately when the instance is turned off.
2. Per-Second Charges for Deployed APIs
For deployed APIs, charges are based on usage during request processing.
How It Works:
You are charged a per-second fee based on the GPU size selected during API deployment.
The fee is incurred only while the API is actively processing a request.
No charges are applied when the API is idle or not in use.
Why This Matters:
This pricing structure is highly efficient, ensuring you only pay for the exact processing time needed.
With InstaSD, you get a fair and scalable pricing model that adapts to your usage.