Agent Credits — Definition
Agent Credits are OpenAI's usage-based billing unit for Workspace Agents, consumed per agent invocation based on model, runtime length, and tool calls.
Longer definition
When the Workspace Agents research preview ended on May 6, 2026, OpenAI shifted to credit-based billing at the workspace level. Each agent invocation consumes credits — the exact amount depends on which model is used, how long the agent runs, how many tool calls it makes, and how much data it processes. Credits are purchased in bundles at the workspace level, with overage alerts and per-agent caps configurable by admins.
In practice
Typical single-run business agents (Weekly Metrics Reporter, Invoice Reviewer, Meeting Prep) cost a fraction of one credit per run. Chatty agents fired hundreds of times per day (Lead Outreach at scale, Support Triage on high-volume channels) cost more. Most SMB deployments land in the $30–$200/month range per active agent; enterprise deployments with dozens of agents and high volume can run into the thousands.
Common mistakes
- Not setting per-agent credit caps — a mis-configured agent can consume unexpected credits
- Comparing credit cost in isolation — the real cost comparison is credits vs human hours saved, not credits vs $0
- Skipping credit estimation during build — size agents for predictable cost rather than discovering cost at end-of-month
See also
Want this built for your team?
20-min intro call. $1,000 per agent, ~1 week to production.