Frequently asked
What counts as a "run"?
A run is a single end-to-end execution of an agent — from trigger to final output. Sub-steps, retries, and tool calls within that run are all included and don't count separately.
Can I use my own LLM provider?
Yes. AgentArc is model-agnostic. Bring your own API keys for OpenAI, Anthropic, Google, or any OpenAI-compatible endpoint. We don't mark up inference costs.
How does self-hosting work?
Enterprise customers can deploy AgentArc into their own AWS, GCP, or Azure environment using our Helm chart. Your data never leaves your infrastructure.
What happens if I go over my run limit?
We'll notify you at 80% usage. On Starter, runs pause until the next billing cycle. On Pro, you can enable overage billing at $0.004 per additional run — or upgrade anytime.
Is there a free trial for Pro?
Yes — 14 days of Pro, no credit card required. After the trial you can continue on Starter or enter payment details to stay on Pro.