Default usage caps for clients
When you create a client via the API, each client is automatically assigned default usage caps for AI features:
| Feature | Default cap |
|---|---|
| Narration tokens | 250,000 / month |
| Image generations | 150 / month |
These caps control how much of your workspace's AI allowance an individual client can consume. They exist to prevent any single client from using a disproportionate share of your allocation.
How do client caps relate to my workspace allowance?
Your workspace has a total AI allowance based on your plan. Client caps sit underneath that. They limit how much each client can use, but all clients draw from the same workspace pool. Your plan-level allowance is a soft cap, so you won't be blocked if total usage exceeds it.
Can I change them?
Not yet. In a future release, you'll be able to adjust per-client caps via the API, giving you full control over how your workspace's AI allowance is distributed across your customers.