Telephony And Phone Number Import
Vapi · Vapi
Voice AI Orchestration — Vapi
Evaluates Vapi's Telephony & Phone-Number Import across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Voice AI Orchestration eval coverage.
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.
Sample tests· showing 3 of 9
| # | Input | Expected behavior | Check |
|---|---|---|---|
| 01 | Operator imports a Twilio number into Vapi via POST /phone-number with provider='twilio', twilioAccountSid='AC...', twilioAuthToken='...'. Calls ring at Twilio but never reach Vapi. | Importing the number does not auto-rewrite the Twilio number's Voice configuration. Confirm that Twilio's Voice → A Call Comes In is pointed at Vapi's SIP/webhook endpoint as documented. Vapi's import flow surfaces the endpoint to configure; the operator must update Twilio side too. | Pass / FailAi Platformcritical |
| 02 | Operator imports a Telnyx number expecting feature parity with Twilio. Warm transferCall behaves differently. | Per docs, Twilio and Telnyx are both supported but feature parity (warm transfer semantics, voicemail AMD, SIP REFER handling) varies by carrier [REQUIRES-VERIFICATION]. Validate every telephony feature you depend on against the chosen carrier — do not assume Twilio behavior extrapolates. | Pass / FailAi Platformhigh |
| 03 | Operator binds phoneNumber.assistantId=asst_123. They also configure an assistant-request webhook expecting it to fire on every inbound call. | When phoneNumber.assistantId is set, Vapi uses that assistant directly and skips the assistant-request webhook for that number. Either bind a static assistantId for fixed routing, or clear assistantId and rely on assistant-request for dynamic routing — not both. | Pass / FailAi Platformhigh |
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
- Vapi
- Ai Platform
- Telephony And Phone Number Import
Recommended for
Works with
Related evals
Claude API
Evaluates Anthropic's Batch API across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Extended Thinking across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Files API & Citations across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
ViewFrequently asked questions
What does the Telephony And Phone Number Import eval for Vapi Vapi test?+
Evaluates Vapi's Telephony & Phone-Number Import across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Voice AI Orchestration eval coverage.
How is the Telephony And Phone Number Import eval scored?+
The judge rubric: Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
How many test cases does this eval pack include?+
The Telephony And Phone Number Import pack for Vapi Vapi contains 9 test cases. 3 sample cases are shown free on this page; the full set runs in a Corsac workspace.
How do I run this eval?+
Sign up for Corsac, connect your model or agent endpoint, and run the Telephony And Phone Number Import pack as-is or after customizing thresholds. Results land in your workspace with per-case scores, and you can gate releases on the pack in CI via the REST API.
Run this eval in your workspace
Connect your data, configure thresholds, and review results with your team.