Rate Limiting

Rate Limit Policy

Effective Date: March 1, 2023

API calls such as aggressive polling or making highly concurrent API calls may result in your app being rate limited. Spectrum’s rate limits may vary based on licensing plan purchased. Rate limiting is a strategy for limiting network traffic. It puts a cap on how often someone can repeat an action within a certain timeframe – for instance, trying to access a Spectrum specification. Rate limiting can help stop certain kinds of malicious activity. It can also reduce load on web servers.

API endpoint limits

To protect platform quality, certain Spectrum endpoints are subject to per-client rate limiting. Today, rate limiting is applied to:

• AI Moderation endpoints (e.g. POST /api/artwork/moderation, POST /api/machine-moderation)
• AI Image Generation endpoints (e.g. POST /api/imagegen/*)

The exact request limit and window are negotiated per licensing plan and are not a fixed public value. Some plans also enforce a separate monthly cap.

Exceeding the configured limit returns an HTTP 429 (Too Many Requests) response with the body Rate limit exceeded. These responses are in addition to any rate-limit headers returned. If your app triggers the rate limit, please refrain from making additional requests until the appropriate amount of time has elapsed.

Requests to rate-limited endpoints must include a valid clientHandle (either as a route value, header, or query parameter, depending on the endpoint). Requests without one are rejected with HTTP 400 (Bad Request) and the body Missing clientHandle.

Exceptions

Please contact your customer service manager to discuss any exceptions needed.

##