Refuel · Rate Limits

Refuel Ai Rate Limits

The Refuel Cloud REST API documents a rate limit of 300 requests per minute and up to 100 concurrent requests against the application label endpoint. Requests beyond these limits are throttled and return HTTP 429. The open-source autolabel library is not rate limited by Refuel; throughput is governed by the underlying LLM provider's limits.

Refuel Ai Rate Limits is the machine-readable rate-limit profile for Refuel on the APIs.io network, conforming to the API Commons Rate Limits specification.

It captures 2 rate-limit definitions, measuring requests and concurrent_requests.

The profile also includes 2 backoff/retry policies defined and response codes documented for throttled.

Tagged areas include AI, LLM, Data Labeling, Data Enrichment, and Autolabel.

2 Limits Throttle: 429
AILLMData LabelingData EnrichmentAutolabelRate LimitingQuotasThrottling

Limits

Requests Per Minute (RPM) account
requests
300
Documented limit of 300 requests per minute on the Refuel Cloud label endpoint.
Concurrent Requests account
concurrent_requests
100
Up to 100 concurrent requests; excess requests are throttled with HTTP 429.

Policies

Throttling
Requests exceeding RPM or concurrency limits return HTTP 429.
Backoff Strategy
Clients should implement exponential backoff with jitter and retry on 429.

Sources