Recharge · Rate Limits

Recharge Payments Rate Limits

The Recharge API is throttled with a leaky-bucket algorithm. Each store token has a bucket that holds up to 40 requests and leaks (empties) at 2 requests per second. A steady rate of 2 requests/second will never trip a limit; short bursts up to the bucket size of 40 are allowed and then refill at 2/second. When the bucket overflows, the API responds with HTTP 429; the guidance is to sleep at least 2 seconds and retry. Higher Recharge plan tiers (Plus, Custom) may be granted higher API rate limits.

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

It captures 4 rate-limit definitions, measuring requests.

The profile also includes 3 backoff/retry policies defined and response codes documented for throttled.

Tagged areas include Subscriptions, Recurring Billing, E-commerce, Rate Limiting, and Quotas.

4 Limits Throttle: 429
SubscriptionsRecurring BillingE-commerceRate LimitingQuotas

Limits

Leaky Bucket Capacity token
requests
40 requests (bucket size)
Maximum number of requests that can be in the bucket at any one instant.
Leak Rate token
requests
2 requests per second
The rate at which the bucket empties; a steady 2 req/sec never trips a 429.
Sustained Throughput token
requests
~120 requests per minute sustained
2 requests/second averaged over time; bursts up to 40 are absorbed by the bucket.
Elevated Tier Limits account
requests
higher on Plus / Custom
Higher Recharge plan tiers may be granted higher API rate limits; confirm with Recharge.

Policies

Backoff Strategy
On a 429 response, sleep for at least 2 seconds before retrying; implement exponential backoff with jitter for repeated overflows.
Bucket Refill
Available request budget increases by 2 for every second in which no request is made, up to the bucket size of 40.
Bulk Operations
Use the async batch endpoints for large create/update workloads rather than issuing many individual requests against the standard endpoints.

Sources