Rate Limits
To keep the platform stable for everyone, we enforce the following limits per API key.
Default limits
| Limit | Value | Scope |
|---|---|---|
| Requests per minute | 10,000 | Per API key |
| Concurrent requests | 50 | Per API key |
| Cost per request | $10 (or 5× balance) | Per request |
| Minimum balance | $0.10 | Per account |
Headers
Successful responses include these headers:
| Header | Description |
|---|---|
X-RateLimit-Limit | Request limit for the window |
X-RateLimit-Remaining | Requests remaining |
X-RateLimit-Reset | Unix timestamp when the window resets |
What happens when I hit a limit?
- Rate limit (429): Wait for the
Retry-Afterheader and retry. - Concurrency limit (429): Finish in-flight requests before starting new ones.
- Balance too low (402): Add credits in your billing settings.
You can create multiple API keys with separate limits to isolate different workloads.