Understanding Usage Credits and billing

CloudRadial AutomationAI meters its own AI usage in Usage Credits, a single unit that covers the AI calls AutomationAI makes on your workspace's behalf. AI that runs on your runner is billed by your own AI provider instead. This article explains what is metered, how AI consumption converts to Credits, what Included Credits cover, and how overage is billed. It is for the Owner and Admin roles.

In this article:

• Credits as the unit of AI cost
• How usage converts to Credits
• Included Credits per period
• Overage and invoicing

Credits as the Unit of AI Cost

AI calls that AutomationAI itself makes consume Usage Credits. These include:

• AI Generate, which drafts workflows, scripts, agents, playbooks, and extensions in AutomationAI
• Knowledge embeddings, which make your knowledge documents searchable, the knowledge searches that ground AI Prompt and Agent nodes, and knowledge consolidation
• Automatic extension selection when the AI Playground chooses extensions for you

AI that runs on your runner does not consume Usage Credits. This covers Agent nodes, AI Playground runs, and the AI calls made by PowerShell, C#, and Python script steps, including AI Prompt nodes. Your runner calls your own AI provider, which bills you directly. The token counts for these steps are still recorded on the run so you can review them.

NOTE: AI that runs on your runner, in script steps, AI Prompt nodes and Agent nodes, is billed by your runner's own AI provider and does not count against Usage Credits.

Credits give you one figure to track instead of reasoning about raw tokens per model. Your current consumption is shown on the Usage page (/settings/usage).

A playbook's Spend limit finish rule is set in estimated credits. It is a cost guardrail, not a bill: it adds up the AI tokens the playbook's workflow steps use on your runners, priced at Usage Credit rates, so you can stop a playbook before it runs up provider costs. Those estimated credits are never deducted from your Usage Credits.

How Usage Converts to Credits

AI usage that consumes Usage Credits is measured in tokens, both the input sent to the model and the output it returns, and converted to Credits using per-model conversion ratios. Token counts recorded for AI that runs on your runner are shown for review only and are not converted to Credits. Each model has its own input and output ratios, so the same number of tokens can cost a different number of Credits on different models. Output tokens are weighted more heavily than input tokens.

The conversion ratios are defined per model in the regional model catalog. The Usage page shows the resulting Credit total, including a breakdown of Credits consumed by AI model, so you can see which models account for your usage.

Included Credits per Period

Each billing period includes a bundle of Credits, the Included Credits. The amount depends on your edition: 1,000 for Starter, 5,000 for Professional, and 5,000 for Enterprise. Trial workspaces start with 10,000. Your edition's Included Credits are shown on both the Usage page and Settings > Overview. Included Credits reset at the start of each billing period.

Overage and Invoicing

Once consumption passes your Included Credits for the period, the additional Credits are counted as overage. Overage Credits are billed; usage within your Included Credits is not.

The current period's Included, Used, and Overage Credit totals are shown on the Usage page. Raw token usage is retained behind the Credit totals so consumption can be recomputed if needed.

If you are still having trouble, we're here to help! Submit a ticket here for assistance, and don't forget to check our status page to ensure there are no outages in your area.