Docs
Rate limits
Token bucket per key. 429s are free and tell you when to retry.
Limits
Each key gets a bucket of 25 requests, refilling at 10 per second. Bursts are fine; sustained traffic settles at 10 rps. Need more? Tell us your volume.
These are the prototype's limits. Production limits will be published here before launch.
Headers
Every response carries X-RateLimit-Limit and X-RateLimit-Remaining.
When you hit it
You get 429 RATE_LIMITED with Retry-After in seconds. It is billed ₹0. Back off, then retry — with an Idempotency-Key so retries are always safe.