Working paper / 2026
Multi-Industry Voice Agent Simulation Bench
Abstract
Multi-Industry Voice Agent Simulation Bench (MIVAS) is an indicator of voice AI performance across economic sectors. MIVAS is comprised of tasks and verifiers with granular rewards executed in production-grade multi-agent RL environments across industries where voice AI is being adopted the fastest. Each task places a model harness inside a multi-agent graph with production-length prompts, tools, and isolated state. Deterministic verifiers score tool use, handoffs, and final database state; pass@1 and passk measure single-run capability and repeated-run reliability.
Clone the repository, pair a model harness with an industry task suite, and reproduce the evaluation environment.
Legal / C1-H2
What this task measures
The caller is on the line with Halverson & Reed for one locked request: Message then premises intake. They keep repeating that ask through a callback offer, a message left for someone else, or advice on a different matter. We included it as a hard reception and routing case because intake has to wait until the caller is classified; messages, wrong numbers, and represented parties are everyday reception work. The agent succeeds only if it stays on that request, reaches the outcome by staying on reception, and leaves the record exactly as this case requires: the right row written or cleared, and nothing extra.
One model harness at a time, with a representative pass and fail conversation when both exist. OpenAI Realtime 2.1 is the default view.
Recording
Conversation cost $0.084
0:00 / 1:23 · audio · trace 803817 · click a span to play, drag to scrub, scroll to pan
Audio
Speech Audio
Agent Audio
Tool Calls
Errors
Transcript
How a conversation passes
Combined pass is the logical and of three deterministic gates. Tools, handoff path, and final database state must all pass on the same conversation. One miss fails the call.
Combined
Fail
This sample fails because tools and final database state did not match. Combined cannot pass unless every gate does.
Expected versus actual tool calls on this conversation.
Expected
Actual
Parameters
full_name: Sam Porter
full_name: Sam Porter
+ phone: 6175550199
Output
+ ok: true
Parameters
for_whom: Tom Eckhart
for_whom: Tom Eckhart
+ message: Sam Porter called about the referral letter. Callback number: 617-555-0199.
Output
+ ok: true
Parameters
- opposing_party: Cedarlane Market
Output
None recorded.
None recorded.
Parameters
- practice_area: premises_liability
Output
None recorded.
None recorded.
Parameters
- practice_area: premises_liability
- state: TX
Output
None recorded.
None recorded.
Parameters
- incident_date: 2026-05-03
- practice_area: premises_liability
- state: TX
Output
None recorded.
None recorded.
Parameters
- incident_date: 2026-05-03
- practice_area: premises_liability
- state: TX
Output
None recorded.
None recorded.
Parameters
+ reason: Message taken and callback promised.
Output
None recorded.
None recorded.
Open a row for parameters and outputs. Required tools must appear. Extra industry tools are allowed. Missing: check_conflict, check_practice_area, check_jurisdiction, calculate_filing_deadline, record_intake. Extra: end_call.
Full industry architecture, same graph as the industry page. The left diagram marks the expected specialist path in green. The right diagram marks the path this conversation actually took in green, because it matched.
Expected path
stay on reception
Actual path
stay on reception
Rows that differ between the expected hangup and the actual hangup.
missing · intakes · 1
Expected
Actual
- caller_id: c_new
- id: 1
- practice_area: premises_liability
- state: TX
- summary: Intake summary.
messages · 1
Expected
Actual
- message: Callback requested.
+ message: Sam Porter called about the referral letter. Callback number: 617-555-0199.
Locked caller script. This is the technical definition. Copy it as written.
Open with exactly: "I need to leave a message for Tom Eckhart." You have two jobs on this call. First they look you up and take a written message for Tom Eckhart. When asked what the message is, say exactly: "Tell him Sam Porter called about the referral letter." Stay until they confirm that message is on the file. Do not hang up after the message. After they confirm the message is written down, say exactly: "I also slipped at a store in Texas and I need a lawyer." This is a new matter of yours. You have not called this firm before. When asked who this would be against, say exactly: "It's Cedarlane Market." When asked which state this happened in or where it took place, say exactly: "In Texas." When asked when, say exactly: "It was May third, twenty twenty six." Give your name and number if asked. Do not ask for a staff member. If offered an appointment or evaluation, decline booking. Stay until they confirm the message is on the file and the intake is recorded. A spoken 'we'll take care of it' without those writes is not enough. Then thank and end. Once they have confirmed the message is on the file and the intake is recorded, say exactly: "Thank you, that's all I needed." and end the call.