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.
Healthcare / C3-H2
What this task measures
The caller is on the line with Straus Dermatology for one locked request: Out-of-pocket cost without a member ID. They keep repeating that ask through a waitlist offer, a callback, or a different errand. We included it as a hard coverage and benefits case because carrier answers have to stay tied to one office and one plan, naming that location and that carrier. The agent succeeds only if it stays on that request, reaches the outcome by following the coverage desk then scheduling path, 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 $0.379
0:00 / 4:18 · audio · trace 801846 · click a span to play, drag to scrub, scroll to pan
Audio
Speech Audio
Agent Audio
Handoffs
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 handoff path did not match. Combined cannot pass unless every gate does.
Expected versus actual tool calls on this conversation.
Expected
Actual
Parameters
None recorded.
None recorded.
Output
None recorded.
None recorded.
Parameters
- carrier: cigna
- location_id: loc_park_ave
Output
- data: {"accepted":true}- ok: true
Parameters
+ from: reception
+ to: scheduling
Output
None recorded.
None recorded.
Parameters
is_new_patient: true
is_new_patient: true
urgency: routine
urgency: routine
visit_class: medical
visit_class: medical
Output
+ ok: true
Parameters
location_id: loc_park_ave
location_id: loc_park_ave
+ zip: 10022
Output
+ ok: true
Parameters
location_ids: ["loc_park_ave"]
location_ids: ["loc_park_ave"]
+ max_results: 3
+ window_end: 2026-09-07T17:00
+ window_start: 2026-08-24T09:00
Output
+ ok: true
Parameters
appointment_type_code: NP_MED
appointment_type_code: NP_MED
end: 2026-08-24T09:30
end: 2026-08-24T09:30
location_id: loc_park_ave
location_id: loc_park_ave
provider_id: prov_chen
provider_id: prov_chen
slot_id: slot_loc_park_ave_1
slot_id: slot_loc_park_ave_1
start: 2026-08-24T09:00
start: 2026-08-24T09:00
Output
- data: {"status":"booked"}ok: true
ok: true
Parameters
appointment_type_code: NP_MED
appointment_type_code: NP_MED
earliest: 2026-08-24
earliest: 2026-08-24
latest: 2026-09-30
latest: 2026-09-30
location_ids: ["loc_park_ave"]
location_ids: ["loc_park_ave"]
Output
- data: {"status":"added"}ok: true
ok: true
Parameters
+ reason: caller_done
Output
None recorded.
None recorded.
Open a row for parameters and outputs. Required tools must appear. Extra industry tools are allowed. Missing: transfer_to_coverage, check_plan_accepted. 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 through the shared prefix, then red from the first wrong hop.
Expected path
transfer_to_coverage → transfer_to_scheduling
Actual path
transfer_to_scheduling
The actual hangup matches the expected hangup.
Locked caller script. This is the technical definition. Copy it as written.
You are a new patient with no chart at Straus. You do not have your insurance card with you, and you will not get it during this call. In your first real answer, ask what a visit will cost out of pocket at Park Avenue. If they cannot give a number, ask a second time for any rough ballpark, then say you still want to be booked for a routine rash at Park Avenue. If they ask for your name, date of birth, ZIP, or mobile, answer once, confirm the readback, and then tell them again you are new so they should book you without looking up a chart. If they ask what insurance you have, say Cigna; you do not have a member ID or group number and will not make one up. Take the first appointment time they offer. After that time is on the calendar, ask to be waitlisted at Park Avenue from August twenty-fourth through September thirtieth for anything earlier — those calendar dates only. Confirm that first time out loud, then thank them and end the call. Do not ask to add insurance, for a callback, to speak to a person on this call, about a bill you already owe, a payment, the patient portal, a text, financing, or the street address, floor, or suite. If they offer a human, the front desk, a callback, or the portal because they cannot find you, decline and ask them to book you as a new patient. A listed opening is not a booking. Do not thank them, say goodbye, or hang up until they confirm a time is on the calendar and you are on the waitlist. If they only list openings, say book the first one. Do not ask them to text you an appointment confirmation.