Rate Limits & Quota

Two independent controls protect the platform and your spend. Both return an OpenAI-compatible error envelope so the standard SDKs surface them cleanly.

Request rate limit (per key)

Each API key is limited to a number of requests per minute (default 60). On overflow you get HTTP 429 with Retry-After plus X-RateLimit-* headers:

HTTP/1.1 429 Too Many Requests
Retry-After: 12
X-RateLimit-Limit: 60
X-RateLimit-Remaining: 0
X-RateLimit-Reset: 12

{"error": {"message": "Rate limit exceeded: this API key allows 60 requests per minute. Retry after 12s.",
           "type": "rate_limit_error", "code": "rate_limit_exceeded"}}

The window is counted per serving replica, so the advertised per-key number is a floor, not an exact global ceiling. Contact Mirobody Support to raise a key’s limit.

Usage quota (per account, monthly)

Each account has a monthly usage cap (USD of billed model usage). When the current month’s usage reaches the cap, /v1 requests are refused with HTTP 429 insufficient_quota until the next month, when it resets:

{"error": {"message": "You have reached your $5 monthly API usage limit (billed $5.00 this month). It resets next month; contact support to raise it.",
           "type": "insufficient_quota", "code": "insufficient_quota"}}

The cap covers all keys under the account together. Cost is metered on provider-true (billed) tokens at the model tier’s rate, with prompt-cache hits billed at the reduced cache rate (see usage.billed_tokens on every response). To raise or lift your cap, contact Mirobody Support.

Client pacing

Path	Suggested RPM	Notes
`POST /v1/chat/completions`	≤ 60	Streaming calls count too
`POST /v1/responses`	≤ 60	Concurrent streams are supported
`POST /v1/files`	≤ 30	Uploads capped at 100 MB per file; OCR is compute-heavy
`POST /v1/data`	≤ 120	Batch records into one request (up to 500/batch)
`GET` reads	≤ 120	—

Client backoff

On 429 rate_limit_exceeded, respect Retry-After — never retry immediately; double the backoff after two consecutive hits, up to 5 minutes.
On 429 insufficient_quota, do not retry — the cap resets monthly or on a support change; surface it to the operator.

​Request rate limit (per key)

​Usage quota (per account, monthly)

​Client pacing

​Client backoff

Request rate limit (per key)

Usage quota (per account, monthly)

Client pacing

Client backoff