How do I request a rate limit increase?

Submit the rate limit increase form. Our team reviews every request, and the more complete your submission, the faster it moves.

Include:

• Model(s) you need the increase on

• Requested RPM and TPM (remember TPM includes input + cached input + output tokens)

• Average input/output tokens per request

• Cache hit rate % — if you don't know it, see below

• Traffic pattern — steady vs. spiky, peak hours and timezone, weekday vs. weekend

• Whether this is for testing/evals or production traffic

More details here usage endpoint