Vespa Cloud pricing is simple and transparent. All customers receive all our features and services, and we charge a fee proportional to the size of the application, summed over three dimensions:
| Per vcpu hour: | $0.11 |
| Per memory Gb hour: | $0.011 |
| Per disk Gb hour: | $0.0004 |
| Per gpu-memory Gb hour: | $0.075 |
This is the only charge and includes everything you need to run production services with confidence:
| 24/7 operations | |
| Support | |
| Continuous deployment | |
| Automatic Vespa/OS upgrades | |
| Strong security | |
| Autoscaling | |
| Performance tune-up program eligibility | |
| Development cloud |
Contact us for priority support, committed resources at lower cost, or other pricing requests.
In production zones you are charged for the resources specified in services.xml for the duration the application is deployed. If the resources are specified as ranges (autoscaling), you are charged for what is actually allocated within the ranges, at each point in time. In dev and test zones, applications will automatically be downscaled to reduce cost, and charges are by the actual downscaled usage summed over time.
In addition, we reserve the right to pass on network ingress/egress costs incurred by application usage initiated explicitly by you, at standard AWS prices, for cases where the network usage is much higher than in a typical Vespa application.
No other costs will be charged.
You can start a free trial for 14 days at any time - no credit card required!