Rate limits
Per-account request and concurrency limits.
Limits apply per account, across all its keys.
| Plan | Requests / min | Concurrent requests |
|---|---|---|
| Pay as you go, Starter | 60 | 5 |
| Pro | 300 | 10 |
| Max | 600 | 16 |
| Status | Code | Meaning | What to do |
|---|---|---|---|
429 |
rate_limited |
Over requests per minute | Wait for retry-after seconds |
429 |
concurrency_limited |
Too many requests in flight | Retry after 1 second, or lower parallelism |
503 |
server_busy |
Service-wide capacity is full | Retry after retry-after seconds (5) |
OpenAI and Anthropic SDKs retry these automatically. Need higher limits? Upgrade your plan.