Select the Correct Budget Control for Your Request Path
Verdictan has more than one budget control. Each control stores a different record and acts on a different request path. This article helps you to select the control that stops your traffic. Usage Costs and Budgets gives the full procedure.
Your model provider bills you directly. Verdictan cost figures are estimates for routing, limits, and analysis.
Start with a token budget
For your first gateway test, set a small API-token budget on a dedicated test token. This budget limits the requests that use that token. It is easier to verify than a billing threshold for the full organization.
Compare the controls
| Control | Configure it in | Request path |
|---|---|---|
| Budget policy | Budget Policies in the Console | Hourly, daily, weekly, or monthly limits for a user, a team, or the organization. The client gets HTTP 402 with usage_authorization_budget_exceeded. |
| Per-minute access limit | limits.rpm_per_token and limits.tpm_per_token in /v1/access-policies | Requests and tokens for each token in a 60-second window. The client gets HTTP 503 with usage_authorization_rate_limited. |
| Billing budget | Budgets in the Console | Cost thresholds, alert settings, and the hard-limit state. Verify that your deployment path applies the gateway block. |
| Scope spend budget | verdictan spend budget or /v1/budgets | Organization, team, user, or key spend. A connected gateway does not load scope budgets for routing. |
| Provider budget | verdictan spend provider-budget | The provider-headroom check for configured gateway traffic. The target must have pricing metadata. |
| Governed API-token budget | API Tokens or verdictan token create --max-budget | The API examines depletion for each request that uses the token. |
A Verdictan budget limits the traffic that goes through Verdictan. It does not limit spend at the provider. If you must have a financial limit, set it in the provider console.
Prove the limit before you rely on it
A saved budget record does not verify enforcement. Do these tests in a test environment:
- Send one allowed request through the gateway.
- Send a request that exceeds the limit.
- Make sure that the gateway blocks the second request.
Read the estimate correctly
Verdictan calculates each estimate from token counts and frozen unit prices. An active administrator price override has priority over the imported model catalog price. If neither price is available, the gateway refuses the request with 503.
A Verdictan amount can be different from the provider amount. The estimate does not include provider discounts, committed-use contracts, taxes, and rounding. Use the provider invoice for accounting.
Next steps
- Usage Costs and Budgets: read cost in Usage and set each budget.
- Spend, Budgets, and API Tokens: manage scope budgets, provider budgets, and token limits from the CLI.
- Compare Model Cost and Latency: compare models before you set a budget threshold.