Skip to main content
Open Usage in the console after making a request. Its Cost, Performance, Tokens, and Activity views show spend, latency, token counts, request details, and prompt-cache hits. The dedicated Safety report covers Guard and leak detection. The Shadow Agent report covers validation outcomes. Your remaining balance appears on the dashboard and Billing page. Each request can include more than one billable component: upstream model inference, SERV reasoning generation, prompt-guard evaluation, audit or repair work, and shadow-agent validation. A cache hit avoids a new SERV reasoning-generation call, but the upstream model still runs and is billed. Test one variable at a time: model, reasoning effort, prompt, tools, schema, or safety feature. Use the same evaluation set to compare task success, failure rate, latency, and total cost. Organization owners can open Billing from the account menu to purchase credits and configure auto top-up. AWS Marketplace organizations see their AWS billing status instead. See Models for model pricing. If your balance cannot cover the request’s estimated ceiling, SERV may reject it before billable inference begins. Use a reasonable output-token ceiling and monitor finish_reason or the endpoint’s stop reason for truncation.