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 / C4-M3
What this task measures
The caller is on the line with Straus Dermatology for one locked request: Chemical peel consult, no deposit. They keep repeating that ask through a waitlist offer, a callback, or a different errand. We included it as a medium cosmetic concierge case because quoted cosmetic policy, deposits, and the cooling-off window have to be spoken before a booking is taken. The agent succeeds only if it stays on that request, reaches the outcome by taking the cosmetic concierge 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 $0.128
0:00 / 1:50 · audio · trace 801931 · 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 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
- service: chemical_peel
Output
None recorded.
None recorded.
Parameters
callback_number: +17185550188
callback_number: +17185550188
+ priority: routine
queue: cosmetic
queue: cosmetic
Output
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_cosmetic, quote_cosmetic_service. 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_cosmetic
Actual path
stay on reception
The actual hangup matches the expected hangup.
Locked caller script. This is the technical definition. Copy it as written.
Ask what a chemical peel costs at Brooklyn Heights. When they explain that a deposit is required to hold the consult, say you are not paying a deposit and you will not agree to that. Ask the cosmetic desk to phone you back later — you do not want to book today and you do not want to be transferred on this call. If they try to connect you to a person now, decline and repeat that you want a callback later, not a live transfer. If they ask for a number to reach you, give 718-555-0188. Stay on the line until they confirm a callback is scheduled, then thank them and end the call. Do not say someone, a person, or transfer me. Do not ask to book, reschedule, or cancel. Do not ask about a medical bill, the patient portal, a text, a waitlist, financing, or the street address, floor, or suite. Do not take a consult time. If they offer a slot, refuse it — you will not put down a deposit and you only want a callback. Do not ask them to text you an appointment confirmation.
