AI Budgets and AI credits serve to measure and control your AI agents usage. Credits are the underlying currency used for every action an AI agent takes, and budgets let you cap how many credits your AI agent(s) can use per calendar month.
What are AI credits?
An AI credit is the unit that an AI agent consumes when it acts on a ticket. Every action an AI agent performs - reading the ticket messages, adding a tag, resolving the ticket, transferring it, etc., is a tool call, and each tool call draws from your account's pool of credits.
The rule is simple: 1 tool call = 1 credit. One MCP tool call made by an AI agent always spends exactly one credit, regardless of which tool it is, or how many input and output tokens the AI consumes. For example, an AI agent run that would use 3 MCP tools to get ticket messages, get ticket notes, and create a knowledge base article would spend 3 credits.
As credits map one-to-one to tool calls, you can estimate the cost of a workflow simply by counting how many actions the AI needs to take per ticket, and multiplying by the number of tickets you expect it to process.
How are AI credits purchased and billed?
AI credits currently do not need to be purchased. To help you get started, every LiveAgent account receives 300,000 AI credits for free. A ready-to-use default budget is also created, with a monthly limit of 300 000 tool calls, so you can put AI agents to work immediately without purchasing anything first.
Since 1 tool call equals 1 credit, 300 000 credits cover 300 000 AI agent actions - enough to run substantial automation across your ticket queue at no additional cost.
What are AI budgets?
An AI budget is a monthly cap/limit on the number of credits (and therefore tool calls) that AI agents may spend.
Budgets allow you to set spending boundaries without having to watch every run individually. You can create as many budgets as needed, and assign each "Run AI agent" rule action to the appropriate one, so you can track and limit spend per AI workflow independently, e.g., per team or workflow. Each budget can be named for easy recognition when assigning it to a rule, and optionally can have a description.
A budget defines the maximum number of credits/tool calls allowed within the given calendar month. Budgets reset on the first day of every month - you do not need to reset them manually.
Budgets are assigned to AI agents via rules that trigger them. When you add the "Run AI agent" action to a rule, you need to select both the AI agent and the budget its tool calls should draw from. Choosing a budget is required - every AI agent run is always tied to a specific budget.
How AI agents spend credits
When an AI agent runs under a budget and makes a tool call, that single call counts against two independent limits at the same time:
- The budget's monthly cap — the call adds 1 to that budget's usage for the current calendar month.
- Your account's credit pool — the call deducts 1 credit from your overall balance.
A run continues making tool calls until it finishes its work — or until it hits one of these limits.
If the assigned budget reaches its monthly cap, or your account's overall credit balance is depleted, the run fails. For more information on AI agent run outcomes, see Monitoring AI agent runs.
Tip: If your AI agent runs frequently fail because the budget's cap is reached, either increase that budget's monthly limit, or simplify your workflows so the AI needs fewer tool calls per ticket. Granting an agent only the tools it actually needs also helps keep spending down.
Monitoring your usage
Under Configuration > AI > AI Budgets you can see your remaining credit balance, the usage of each budget for the current month, and an estimated credit runway — roughly how many days of AI agent activity your remaining credits will cover at your current budgeted rate. Reviewing this regularly helps you catch a workflow that is spending faster than expected before it runs out.