Platform Rate Limits: Buy and Order
We are introducing rate limits on Platform purchase and quote requests to improve stability and the Platform experience. Rate limits exist at the client level, meaning Platform Web App, SDK, and direct API activity share the same limits.
Limits
All clients can make up to:
- 100 requests per minute
- 1,000 requests per hour
Both limits apply simultaneously. Each window begins with you first request and resets when that window closes.
How Rate Limits Apply
Enterprise: Rate limits are shared at the client level. All users and workflows, including the Platform UI, API, and SDK, draw from the same overall limits of 100 requests per minute and 1000 per hour. Please contact your account manager with any questions about your limits.
Pro and Business: Each user has an individual limit of 100 requests per minute and 1,000 requests per hour.
If you reach a limit, the Platform UI and API will tell you how long to wait before requests can resume.
How Rate Limit Windows Reset
Rate limit windows begin with your organization's first request, not at the top of the minute or hour. Once started, the window remains fixed and does not extend with additional requests.
For example, if your first purchase request starts an hourly window at 5:37 PM, that window resets at approximately 6:37 PM, not 6:00 PM.
If you reach a limit, the Platform UI and API will tell you how long you need to wait before the window resets and requests can resume.
What's Covered
The limits currently apply to purchase and quote requests across the below in our API and Web App:
framework/order: creating an orderframework/buy: purchasing a framework
If You Reach a Limit
Requests above either limit will return a 429 Too Many Requests response with a Retry-After value indicating when you can resume.
For automated workflows, wait for the specified Retry-After period before retrying. Avoid immediately retrying requests, as all users and workflows within your organization share the same limits.
Guidance for API and SDK Users
For larger automated workflows, we recommend batching requests where possible and managing throughput against both the minute and hourly limits. As a reference, 1,000 requests per hour is the maximum sustained volume, even if individual minutes remain below 100 requests.
Because limits are shared at the client level, parallel workers, Platform activity, SDK usage, and direct API requests all contribute to the same limit.
Questions
If these limits impact your workflow or you expect to require higher-volume purchasing, please contact your Carbon Arc account team or support@carbonarc.co.