ezgh inference limits

ezgh inference limits

Show the limits in force for a project (--project, EZGH_PROJECT or the profile’s): requests per minute for the project, and input tokens per minute, concurrent requests and input tokens per UTC month for the organization; then each model’s limits per request. Over a rate, the API answers 429 with Retry-After; over the monthly tokens, 429 insufficient_quota.

Part of ezgh inference.

Flags

This command also takes the global flags.