>
    Strata Copilot
    Configure Budget Limits
    Focus
    Download PDF
    Focus
    1. Home
    2. Prisma AIRS
    3. AI Gateway
    4. Add an LLM Integration
    5. Configure Budget Limits
    Download PDF
    Prisma AIRS

    Configure Budget Limits

    Table of Contents

    Configure Budget Limits

    Set a cost-based or token-based spending cap on a workspace's LLM integration to prevent unexpected charges and control flex credit consumption.
    Where Can I Use This?What Do I Need?
    • Prisma AIRS AI Gateway (Americas region)
    • AI Gateway activated
    • LLM integration in progress or saved
    Budget limits cap the total spending for a workspace on a specific LLM integration, either in USD cost or in token count. AI Gateway tracks consumption per request and stops routing traffic from the workspace once the limit is reached. Budget limits complement rate limits: rate limits control request velocity, while budget limits control total spending over a period. When a workspace exhausts its budget, AI Gateway returns HTTP 412 (Precondition Failed) until the budget resets or is raised.
    AI Gateway tracks flex credit consumption at 1 flex credit per 4 characters of LLM input or output. When you set a token-based budget, the limit is applied to the total token count across all model calls from that workspace.
    1. During LLM integration setup, select the workspace row for which you want to set a budget limit.
    2. For Budget Type, choose Cost (USD) or Tokens.
    3. In the budget amount field, enter the spending cap.
      Minimum values: $1 USD for cost-based budgets, or 100 tokens for token-based budgets.
    4. For Reset Period, choose how often the budget resets: No Reset, Daily, Weekly, or Monthly.
      Daily budgets reset at 00:00 UTC. Weekly budgets reset on Monday at 00:00 UTC. Monthly budgets reset on the first day of each calendar month at 00:00 UTC. If you select No Reset, the budget is a lifetime cap and requests are blocked permanently once the limit is reached until you manually raise it.
    5. Repeat for each workspace that requires a budget limit.
      When a workspace exhausts its budget, AI Gateway returns HTTP 412 (Precondition Failed) for all subsequent requests until the budget period resets. Budget consumption and limit events are visible in the observability logs.

    © 2026 Palo Alto Networks, Inc. All rights reserved.