1. How do I manage the concurrency that I want my ai assistant to dial?... The dashboard in the billing section won't let me lower it. 2. I just uploaded a list of 500 leads for the ai assistant to call and it seems like it has stopped calling. When I checked the call logs, I noticed some errors messages, specifically two that stated: "An error occurred in this call: call.in-progress.error-providerfault-outbound-sip-503-service-unavailable" and "An error occurred in this call: pipeline-error-playht-request-timed-out". 3. Where can I see each ai assistant's call queue? 4. How do I setup local presence dialing within Vapi dashboard? How do I resolve these issues?... I'm currently using Vapi as the telephony provider and LLM: OpenAI with GPT 4o Mini Cluster model and Voice Configuration: Playht using its Playht2.0 model.

Batch your 500 calls to stay within Vapi’s 10-call concurrency limit, add retry logic and a fallback voice like OpenAI’s "shimmer" to handle PlayHT timeouts, throttle outbound calls to prevent SIP 503 errors, use webhooks or the Vapi API to monitor assistant call queues, and enable local presence dialing by assigning local numbers or SIP trunks in the dashboard..

How do I batch my list of 500 leads?… How do I throttle outbound calls? Where do I find the documentation for all of the solutions you provided?

Upload them via API or CSV, Vapi will process up to your concurrency limit and queue the rest automatically, throttle outbound calls using built-in queuing, webhooks, or the List Calls API.. https://docs.vapi.ai/calls/outbound-calling

Hey @Billions14, checking if this is resolved/solved for you?