Rate Limits
To prevent abuse, Voice AI enforces rate limits on the public endpoints. How you authenticate decides the ceiling you hit.
Public vs. authenticated
| Access | Limit | Best for |
|---|---|---|
| Public /public endpoints | Capped per IP, per day | Prototyping & demos |
| Authenticated (bearer token) | Higher, account-level limits | Production traffic |
Non-public endpoints need an Authorization header with a bearer token. Get one from
the Authentication guide.
Handling 429
When you exceed a limit, the API responds with 429 Too Many Requests. Back off and
retry, or switch a public call to an authenticated one for a higher ceiling.
{
"errors": {
"message": "The daily limit of AI model requests for your IP address has been reached.",
"status": 429
}
}Tips
- Prefer authenticated endpoints in production. Public limits are meant for trying things out.
- Implement exponential backoff with jitter when you receive
429or5xx. - Keep audio files within the supported size. Oversized files return
400.