Rate limits

Rikaii applies distributed rate limits to POST /v1/chat/completions so shared infrastructure stays reliable for everyone.

Limits (requests per minute)

Limits are enforced per workspace (identified by your API key) using a one-minute sliding window:

Plan	Requests per minute (RPM)
Standard (pay-as-you-go, non‑Pro)	60
Pro	120

These values match the deployed rate-limit configuration. If you need sustained higher throughput, contact support.

HTTP behavior

When the limit is exceeded, the API returns 429 Too Many Requests with a JSON error body. Rikaii does not send a Retry-After header—use exponential backoff when you receive 429.

If rate-limit checks are temporarily unavailable, the API may return 503 Service Unavailable instead of allowing unlimited traffic (fail-closed).

Best practices

• Prefer backoff and jitter on 429 rather than tight retry loops.
• Serialize bursty traffic within your tier limits.

Related

• Errors
• Authentication