Sume API rate limits by plan: requests per minute for writes and reads

The Sume API gives each key a request budget per minute across all of /v1, set by the plan of the workspace the key belongs to: 120 writes a minute on Free, 300 on Pro, 600 on Startup and 1200 on Scale. Reads have their own bucket at forty times the write number, so a tight status-poll loop cannot 429 your own submits.

The numbers

A read is any GET or HEAD, plus two POSTs that submit nothing. Everything else, including creating runs, cancelling and uploads, is a write. Enterprise is contract-based; until a number is provisioned an Enterprise key resolves to the Scale row.

Do not confuse it with generation capacity

Request rate is a different control from how many generations run at once. Concurrency and queue capacity come from your plan's generation_limits, and raising your request rate does not raise them. A 429 can mean either: rate_limited is the request budget, queue_full is the generation queue.

Read the headers instead of counting

Responses carry ratelimit-limit, ratelimit-remaining, ratelimit-reset and, on a 429, retry-after. They describe whichever bucket the current request spent from, and a 429 names it in error.details.scope as read or write. The read multiple is a deployment setting, so the headers on the response are the authority for the host you call, not this table.