API
Your systems, our gateway.
Call your LLM profiles from your own systems: every key has its own usage, metered per call.
OpenAI compatible
The same call format you already use. The « model » field names a profile of your space.
The provider never leaks
You call a profile, not a provider. What runs behind it can change without touching your code.
One key, one usage
Every key has its own counter: calls, tokens, cost. You revoke the one that runs away without cutting the others.
- 01
You create a key from your space.
- 02
You call the endpoint with the profile you want.
- 03
Usage shows up key by key.
Can the secret be retrieved?
No. It is shown once, at creation. Lost, it gets revoked and replaced, never read again.
Is there a cap?
Your space's, in euros. A runaway key stops at the cap instead of running the bill away.
The other modules
Ready to get that time back?
We'll show you ERHA on one of your own files, in 30 minutes.