Billing and resource consumption are calculated on a per-minute basis. There is no minimum resource charge.
For Pods/Instances with a GPU attached, the rate of billing is determined by the price of the GPU per hour. The GPU price per hour, as listed on the CoreWeave Cloud Pricing page, includes the host device vCPU, RAM and ephemeral storage allocations.
CPU Instance Billing
For Pods/Instances without a GPU attached, the rate of billing is determined by the number of vCPU cores requested, or the current actual vCPU core usage, whichever is higher.
All CoreWeave Cloud billing periods cover the calendar month (i.e. 1 January 12:00am UTC thru 1 February 12:00am UTC).