For VapiAI Platform

Batch Calls And Concurrency

Vapi · Vapi

Voice AI Orchestration — Vapi

Vapi evals — Batch Calls & Concurrency (relift v3 InfraRed)

About Vapi

Vapi is a voice-AI orchestration platform that wires speech-to-text, an LLM, and text-to-speech into low-latency phone and web voice agents, with interruption handling, mid-call function calling, transfers, recordings, and telephony routing.

Employees

~50

Industry

Voice AI Orchestration

Headquarters

San Francisco, CA

Website

vapi.ai

Sample tests· showing 3 of 9

#	Input	Expected behavior	Check
01	Operator dials 5000 leads by issuing 5000 POST /call requests in a tight loop. First batch succeeds; later calls fail with 429.	Vapi has tenant-level concurrency / RPS caps [REQUIRES-VERIFICATION on numeric values]. Pace /call creation to stay under the cap, queue overflow operator-side, and back off on 429 with documented retry-after. Track in-flight call count from status-update events.	Pass / FailAi Platformhigh
02	Operator counts in-flight calls by 'POST /call returned 200' minus 'end-of-call-report received.' Counts drift over multi-day campaigns.	Track in-flight from status-update (queued → ringing → in-progress → ended). Reconcile by call id, not by event counts. Persist last-seen status per call; scrub stale 'in-progress' calls older than max call duration via the REST API to detect missed end-of-call-report deliveries.	Pass / FailAi Platformmedium
03	Customer doesn't pick up. Operator wants to retry tomorrow at the same time.	Inspect artifact.endedReason — 'customer-did-not-answer' / 'voicemail-left' / 'busy' have different retry semantics. Schedule retries operator-side (Vapi is not a campaign manager). Respect carrier and TCPA call-window rules on retry timing.	Pass / FailAi Platformcritical
Use this eval