Retell Function Calling Custom Tools
Retell AI · Retell AI
Voice AI Agents — Retell AI
Evaluates Retell AI's Function Calling / Custom Tools across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Voice AI Agents eval coverage.
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.
Sample tests· showing 3 of 9
| # | Input | Expected behavior | Check |
|---|---|---|---|
| 01 | Agent defines a custom function check_appointment with a parameter schema, but omits a clear description, so the model invokes it with wrong/empty arguments. | Define the custom function with a precise name, description, and parameter schema (types + required fields) so the model knows when and how to call it. Validate arguments server-side before executing; ambiguous descriptions cause mis-invocation. | Pass / FailAi Platformhigh |
| 02 | A custom function's backend hangs past its timeout. The agent waits indefinitely with the caller on the line. | Set a function timeout and, on expiry, return a graceful fallback to the model ('I couldn't reach that system; can I take a message?') rather than hanging. Bound the wait so the call stays responsive. | Pass / FailAi Platformcritical |
| 03 | A custom function (CRM lookup) takes a few seconds. During execution the agent goes completely silent and the caller thinks the line dropped. | Configure the function to speak a brief filler/execution message ('Let me pull that up') while the backend runs, then deliver the result. Long silent gaps on a voice call read as a dropped call; cover the latency audibly. | 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
- Retell
- Ai Platform
- Function Calling Custom Tools
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 Retell Function Calling Custom Tools eval for Retell AI Retell AI test?+
Evaluates Retell AI's Function Calling / Custom Tools across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Voice AI Agents eval coverage.
How is the Retell Function Calling Custom Tools 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 Retell Function Calling Custom Tools pack for Retell AI Retell AI 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 Retell Function Calling Custom Tools 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.