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 / C4-M3
What this task measures
The caller is on the line with Halverson & Reed for one locked request: Packet by text message. 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 medium intake and documents case because taking a narrative and a document packet is representative plaintiff work that starts after the conflict-before-facts rule has held. The agent succeeds only if it stays on that request, reaches the outcome by following the conflict screening then intake 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 $0.333
0:00 / 4:13 · audio · trace 804021 · 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
Pass
All three gates matched, so the conversation counts as a pass.
Expected versus actual tool calls on this conversation.
Expected
Actual
Parameters
+ full_name: Priya Nair
+ phone: 4085550199
Output
+ ok: true
Parameters
+ from: reception
+ to: screening
Output
None recorded.
None recorded.
Parameters
opposing_party: Apex Collections
opposing_party: Apex Collections
Output
+ ok: true
Parameters
practice_area: consumer
practice_area: consumer
Output
+ ok: true
Parameters
practice_area: consumer
practice_area: consumer
state: CA
state: CA
Output
+ ok: true
Parameters
incident_date: 2026-02-10
incident_date: 2026-02-10
practice_area: consumer
practice_area: consumer
state: CA
state: CA
Output
+ ok: true
Parameters
+ from: screening
+ to: intake
Output
None recorded.
None recorded.
Parameters
incident_date: 2026-02-10
incident_date: 2026-02-10
practice_area: consumer
practice_area: consumer
state: CA
state: CA
- summary: Debt collector harassment at work.
+ summary: Caller reports a debt collector repeatedly called her at her workplace starting on February 10, 2026. She describes the calls as stressful and embarrassing. She prefers not to book a case evaluation at this time and requests the new client packet by text.
Output
+ ok: true
Parameters
channel: sms
channel: sms
Output
+ ok: true
Parameters
+ reason: Intake recorded, packet sent by sms, caller declined to book evaluation at this time.
Output
None recorded.
None recorded.
Open a row for parameters and outputs. Required tools must appear. Extra industry tools are allowed. Parameter miss: record_intake. Extra: lookup_caller, transfer_to_screening, transfer_to_intake, 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
transfer_to_screening → transfer_to_intake
Actual path
transfer_to_screening → transfer_to_intake
The actual hangup matches the expected hangup.
Locked caller script. This is the technical definition. Copy it as written.
Open with exactly: "A debt collector keeps calling me at work." When asked who this would be against, say exactly: "It's Apex Collections." California. February tenth, twenty twenty six. If they ask whether this is employment, workers' compensation, or a job claim, say exactly: "No. It's a debt collector, not my employer." If offered the packet, say exactly: "Text is fine." Decline booking. Stay until they confirm the intake is on the file and the packet was sent by text. A spoken promise without those writes is not enough. Once finished, say exactly: "Thank you, that's all I needed." and end the call.
