🌐 เอกสารภาษาไทยกำลังจัดทำ — เนื้อหาด้านล่างเป็นภาษาอังกฤษชั่วคราว จนกว่าจะมีการแปล. This page is not yet translated; English content is shown temporarily.

Usage & budget

The Overview dashboard shows what you've used and what you have left — so there are no billing surprises.

Prerequisite

You're signed in and your account is linked to a consumer (Get access).

What you see

• Monthly budget — your USD budget, the amount spent, and the amount remaining for the current calendar month (UTC).
• Token usage — input, output, and cached tokens.
• Per-model breakdown — cost by model, so you can see where your spend goes.

How budgets work

Your budget is part of a hierarchy — organization ≥ project ≥ group ≥ user — and the tightest cap applies. When you reach a cap, further requests are rejected with 429 Too Many Requests until the next month or until an administrator raises the limit.

Semantic caching can stretch your budget: repeated or similar prompts may be served from cache, which costs nothing against your budget.

If you hit your limit

• Wait for the monthly reset, or
• Ask your org admin to adjust your budget or limit.

Next steps

• Blocked requests — if a request was rejected by a guardrail rather than a budget.
• Request lifecycle — which gate produced a given status code.