Eval Library
RA
For Retell AIAI Platform

Retell Telephony And Call Lifecycle

Retell AI · Retell AI

Voice AI Agents — Retell AI

Retell AI evals — Telephony & Call Lifecycle (relift v3 InfraRed)

About Retell AI

Retell AI is a platform for building production phone-call voice agents — pairing a conversation engine with telephony, low-latency turn-taking, interruption handling, mid-call functions, post-call analysis, and batch outbound dialing.

Employees

~40

Industry

Voice AI Agents

Headquarters

San Francisco, CA

Sample tests· showing 3 of 9

#InputExpected behaviorCheck
01

Integrator uses custom telephony and tries to dial the SIP URI for an inbound call without first calling Register Phone Call.

Call Register Phone Call first to bind the agent and obtain the call_id used to construct the SIP URI; only then dial. The registration tells Retell which agent handles the call. Skipping it leaves Retell without the agent binding.

Pass / FailAi Platformcritical
02

After Register Phone Call, the dialer is delayed in a queue and connects the SIP call ~8 minutes later.

Dial within the registration validity window (docs state ~5 minutes); a late connection disconnects with registered_call_timeout. Re-register if the dial will be delayed. [REQUIRES-VERIFICATION on the exact current window duration.]

Pass / FailAi Platformhigh
03

A phone number is bound to one agent, but the operator wants different agents for inbound vs outbound on that number.

Configure inbound and outbound agents separately (Retell supports distinct inbound/outbound agents, with the option to disable either). Do not assume one agent field covers both directions; verify each direction routes to its intended agent.

Pass / FailAi Platformmedium

How this eval is graded

Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.

Rubric criteria

  • Retell
  • Ai Platform
  • Telephony And Call Lifecycle

Recommended for

Retell AIRetell AI customers

Works with

Related evals

Run this eval in your workspace

Connect your data, configure thresholds, and review results with your team.