Monitoring usage metrics and costs for Vultr Serverless Inference subscriptions through the Customer Portals Usage tab
You can monitor the usage and cost of your Vultr Serverless Inference subscription through the Vultr Customer Portal. The Usage tab provides detailed information about token consumption, including the total number of input and output tokens processed. This allows you to track how workloads translate into usage and monitor spending in near real time.
Serverless Inference pricing is based on token consumption. Requests are billed at $0.55 per 1,000,000 input tokens and $2.75 per 1,000,000 output tokens. The Usage tab reflects the total tokens consumed, helping you understand how your inference workloads affect billing. Visit for more pricing information.
For a broader overview, the Overview tab displays subscription details such as your API key and API status, allowing you to compare service configuration with your current usage patterns.
For more details, see How to Monitor Vultr Serverless Inference, which also explains how to retrieve usage data via the Vultr API and CLI.