Forecasting
Forecasting APIby Birla AI Labs

Rate Limits

Limits protect the service and are enforced per API key and per account. Requests over a limit receive an HTTP 429 response.

Limits

LimitValueScope

Request rate

Sliding 60-second window. Configurable per key.

60 / minper API key

Daily quota

Rolling 24-hour window.

200 / dayper account

Weekly quota

Rolling 7-day window.

1,000 / weekper account

Series per request

Maximum time series in one forecast call.

100per request

The maximum forecast horizon depends on the model — see the per-model limits in Available Models.

When a limit is exceeded

429Request rate exceeded
{ "detail": "Rate limit exceeded. Please try again later." }
429Daily quota exceeded
{ "detail": "Daily API request limit exceeded. Limit: 200 requests per day." }
429Weekly quota exceeded
{ "detail": "Weekly API request limit exceeded. Limit: 1000 requests per week." }

Handling 429s

On a 429, back off and retry with exponential delay (for example 1s, 2s, 4s, 8s) plus a little random jitter. Spread bulk workloads across time rather than bursting, and batch multiple series into a single request (up to 100) instead of sending many small calls.