API Rate Calculator
Convert between per-second, per-minute, per-day API rate limits and model burst capacity.
Free & unlimited
Preset
Rate
Time window
Burst window (s)
Burst multiplier
Per second
10.00
Per minute
600.00
Per hour
36.00K
Per day
864.00K
Burst max
200.00
in 10s
Refill rate
10.00
tokens/s
Monthly cap
25.92M
requests
Rate over 60 seconds
0s30s60s
All processing happens in your browser. No data is sent to any server.
About this tool
- 1
Enter your rate limit
Input the allowed number of requests and the time window (per second, minute, or hour).
- 2
Set your usage
Enter your expected request volume or average requests per time period.
- 3
View the breakdown
See the calculated rates converted across all time units and your usage percentage.
- 4
Plan for scale
Adjust projected growth to see when you will hit the rate limit.
- Convert between per-second, per-minute, and per-hour rates to compare different API providers.
- Factor in retry requests when estimating your actual API usage - retries count toward the limit.
- Use the burst calculation to understand how many concurrent requests you can send safely.
- Add a 20% safety margin to your estimates to account for traffic spikes.
- Converts rate limits between per-second, per-minute, per-hour, and per-day units
- Usage percentage calculator showing how close you are to the limit
- Burst rate and sustained rate calculations
- Growth projection showing when you will exceed the limit at current trends
- Cost estimation for overage charges based on provider pricing
- Determine if an API pricing tier meets your application throughput requirements
- Calculate how many parallel workers you can run without exceeding rate limits
- Compare rate limits across different API providers when evaluating alternatives
- Plan capacity by projecting when current usage will exceed the allowed rate
Burst rate is the maximum requests per second you can send in short spikes. Sustained rate is the average over a longer window like one minute or one hour.
Implement exponential backoff: wait 1 second after the first 429, then 2, 4, 8 seconds, etc. Check the Retry-After header for the exact wait time.
Yes. Every request sent to the API counts, including retries. Factor retries into your usage estimates.