Rate Limits

Limits by plan

Each HypeRPC plan has defined limits for monthly compute units (CU) and requests per second (RPS). Exceeding the RPS limit returns a 429 status code. Monthly CU overages are billed separately.

Compute units

Not all RPC methods cost the same. The Console dashboard shows your CU usage in real time.

Handling 429 responses

When you exceed your RPS limit, HypeRPC returns HTTP 429 Too Many Requests. The response includes a Retry-After header indicating how many milliseconds to wait before retrying.

Recommended retry strategy:

1. On receiving a 429, wait for the duration specified in Retry-After

2. If no header is present, use exponential backoff starting at 100ms

3. Cap the backoff at 5000ms maximum

4. Add random jitter (0–50ms) to avoid thundering herd issues

Example backoff sequence: 100ms → 200ms → 400ms → 800ms → 1600ms → 3200ms → 5000ms (cap).

Burst allowance

All plans include a short burst allowance of 2× your RPS limit for up to 3 seconds. This handles brief traffic spikes without returning 429 errors. Sustained traffic above your base RPS will trigger rate limiting after the burst window.

Additional compute units

If you need more monthly CU without upgrading your plan, you can purchase additional blocks. Configure overage billing in the Console under Billing → Usage Alerts.

WebSocket limits

WebSocket connections have separate limits. Each connection can subscribe to up to 50 channels simultaneously — messages are not rate-limited.