Rate limits
deepface.dev applies rate limiting and queue controls at the gateway.Enforcement points
- Per-account requests per minute
- Maximum inflight compute requests
- Queue depth
- Queue timeout
Retry behavior
429 rate_limited responses include Retry-After in seconds.
503 queue_full and 503 queue_timeout indicate temporary capacity pressure.
Treat them as retryable with backoff unless your own SLA policy says otherwise.