Features / Rate limiting
LLM API rate limiting for every agent on your team.
Provider rate limits protect the provider, not your budget. ACPpotamia adds your own requests-a-minute limit in front of OpenAI, Anthropic and self-hosted models, so a burst from one agent cannot exhaust your quota or your money.
How it works.
Per team, key, customer and session
A sliding one-minute window for the whole team, plus optional limits per API key, per customer (X-ControlPlane-Tenant-Id) and per session, all checked at the edge before the call leaves.
Standard 429 with Retry-After
SDKs already know how to back off; the response says how many seconds to wait.
Up to 10,000 requests a minute
20 on Free, up to 600 on Pro, up to 10,000 on Enterprise. You choose the number below your plan's ceiling.
Questions.
How is this different from OpenAI's rate limits?
OpenAI limits protect OpenAI's capacity. ACPpotamia's limit is yours: it protects your budget and your quota from one runaway agent.
Do blocked requests cost anything?
No. They are refused before reaching the provider, and they still appear in the audit log.
Related features.
Budgets and cost control
Put a hard monthly budget on OpenAI and Anthropic spend. ACPpotamia reserves each AI agent call's cost before it is sent, so agents cannot overspend.
Loop detection
Detect and stop AI agents stuck in loops: repeated tool calls, retries and identical prompts. ACPpotamia stops the agent in seconds, before it burns your budget.
Kill switch
Stop one AI agent session or every agent on your team with one click or one API call. The kill switch takes effect on the next call, everywhere.