Rate Limits

Alpaca applies rate limits to the Broker API to protect the stability and fairness of the platform for every partner. This page explains how to detect where you stand, what happens if you exceed your limit, and how to build an integration that handles limits gracefully.

How limits are applied

Broker API rate limits are applied at the correspondent level - that is, across your whole integration, not per end-user account. Every partner starts with a default limit, and limits can be tailored based on demonstrated, sustained usage. Because limits are set per partner and reviewed case by case, Alpaca does not publish fixed Broker API rate-limit numbers.

Rate-limit headers (returned on every response)

Every Broker API response - not only 429 responses - includes headers that tell you exactly where you stand in the current window:

Best practice: monitor X-RateLimit-Remaining proactively and slow down before it reaches zero, rather than waiting to be throttled.

What happens when you exceed the limit

When you exceed your limit, the request returns HTTP 429 Too Many Requests. When you receive a 429, stop, wait, and retry using exponential backoff. Ideally, do not retry before the time indicated by X-RateLimit-Reset.

Exponential backoff

Instead of retrying a throttled request immediately (which only adds load and produces more 429s), wait a short interval and double it after each subsequent failure - e.g. 1s, then 2s, 4s, 8s - up to a maximum, until the request succeeds. Add a small amount of randomness ("jitter") to each wait so that many clients don't retry in lockstep.

Example (pseudocode):

delay = 1.0          # seconds
max_delay = 60.0
for attempt in range(max_retries):
    response = send_request()
    if response.status_code != 429:
        break
    sleep(delay + random.uniform(0, 0.5))   # add jitter
    delay = min(delay * 2, max_delay)       # exponential backoff

If you need a higher limit

Increases are reviewed based on demonstrated need. If you have a genuine, sustained requirement for a higher limit, reach out to Alpaca and be specific about the driver: your use case, expected peak requests per minute, and the time pattern of the traffic. Limits exist to keep the platform stable for everyone, so increases are granted against real, evidenced usage.