Working paper / 2026
Conversational AI LLM Bench
Abstract
We define conversational AI LLMs as the language models at the center of cascaded voice pipelines, between speech recognition and speech synthesis, that return a first token within one second: a median time to first token of 1,000 ms or less. The Conversational AI LLM Bench (CALLM) benchmarks this category. Each model operates every agent of a multi-agent front desk in healthcare, legal, and customer support, with production-length prompts, executable tools, and per-call database state. A simulated caller pursues a locked task over a simulated phone line, so the model receives only what a speech recognizer would have transcribed, errors included. A deterministic verifier scores tool calls, handoffs, and final database state; pass@1 and pass5 measure single-attempt capability and five-attempt reliability.
Clone the repository, point a model at an industry pack, and reproduce the evaluation environment.
Healthcare / C4-H4
What this task measures
The caller is on the line with Straus Dermatology for one locked request: Existing patient Botox after Windermere cannot do cosmetic. They keep repeating that ask through a waitlist offer, a callback, or a different errand. We included it as a hard 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 following the identity then cosmetic concierge path, and leaves the record exactly as this case requires: the right row written or cleared, and nothing extra.
One model at a time, with a representative pass and fail conversation when both exist. GPT-4.1 is the default view.
Conversation cost $0.131
0:00 / 0:00 · scrub · trace sample · 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 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
next_intent: cosmetic
next_intent: cosmetic
Output
+ role: identity
+ success: true
Parameters
- dob: 1995-09-08
- first_name: Alice
- last_name: Romano
Output
None recorded.
None recorded.
Parameters
- dob: 1995-09-08
- full_name: Alice Romano
Output
None recorded.
None recorded.
Parameters
None recorded.
None recorded.
Output
None recorded.
None recorded.
Parameters
None recorded.
None recorded.
Output
None recorded.
None recorded.
Parameters
- service: botox
Output
None recorded.
None recorded.
Parameters
location_id: loc_windermere
location_id: loc_windermere
Output
+ data: {"locations":[{"id":"loc_windermere","name":"Windermere","zip":"34786","offers_cosmetic":false,"address":"7600 Conroy Windermere Road, Windermere, FL 34786","floor":"Ground floor","suite":"Suite 100","hours":"Mon-Fri 8am-5pm","services":"medical, surgical, allergy","transit":"No transit nearby","parking":"Free surface lot"},{"id":"loc_brooklyn_heights","name":"Brooklyn Heights","zip":"11201","offers_cosmetic":true,"address":"142 Montague Street, Brooklyn, NY 11201","floor":"2nd floor","suite":"Suite 2B","hours":"Mon-Fri 8am-5pm","services":"medical, surgical, cosmetic, allergy","transit":"Two blocks from Borough Hall","parking":"Street parking; garage on Clinton Street"},{"id":"loc_park_ave","name":"Park Avenue","zip":"10016","offers_cosmetic":true,"address":"386 Park Avenue South, New York, NY 10016","floor":"4th floor","suite":"Suite 410","hours":"Mon-Fri 8am-5pm, Sat 9am-1pm","services":"medical, surgical, cosmetic, allergy","transit":"One block from the 6 train at 33rd Street","parking":"Garage next door on East 27th"}]}+ error_code: empty
+ ok: true
+ patient_safe_message: empty
Parameters
- location_ids: ["loc_park_ave"]
Output
None recorded.
None recorded.
Parameters
- location_id: loc_park_ave
- policy_acknowledged: true
- provider_id: prov_chen
- service_interest: ["botox"]
- slot_id: slot_loc_park_ave_1
- start: 2026-08-24T09:00
Output
None recorded.
None recorded.
Parameters
None recorded.
None recorded.
Output
None recorded.
None recorded.
Parameters
- mobile_e164: +14075550155
Output
None recorded.
None recorded.
Open a row for parameters and outputs. Required tools must appear. Extra industry tools are allowed. Missing: identify_patient, verify_identity, get_patient_summary, transfer_to_cosmetic, quote_cosmetic_service, find_slots, book_cosmetic_consult, offer_financing, send_payment_link.
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_identity → transfer_to_cosmetic
Actual path
transfer_to_identity
Rows that differ between the expected hangup and the actual hangup.
missing · appointments · 28
Expected
Actual
- appointment_type_code: COS_CONSULT
- description: Cosmetic consult: botox
- end: 2026-08-24T09:30
- id: 28
- location_id: loc_park_ave
- patient_id: pat_alice_romano
- provider_id: prov_chen
- start: 2026-08-24T09:00
- status: booked
The locked caller script. Copy it as written.
You are an existing Straus patient in Florida. Answer identity questions once with your name and date of birth, confirm if they read it back, then get to why you called. After they greet you or ask how they can help, your first spoken request is how much Botox costs and whether Windermere does cosmetic work; do not open by saying you still need a consult booked. Ask how much Botox costs, then ask them to look up whether Windermere does cosmetic work. When they say Windermere does not do cosmetic work, ask for a consult at Park Avenue instead. You need the Park Avenue street address, the floor, and the suite before you will take any time; if they skip any of those, ask again. After you hear the Botox price, say that is more than you can pay all at once and ask whether there is a way to spread the treatment cost out. When they explain the deposit and cancellation rules, agree. Ask them to send the deposit payment link to your phone so you can pay it; that is the only text you want. After you have the address, take the first appointment time they offer. Do not ask for a different day, time, or office, and do not suggest any date of your own. Confirm that first time out loud, then thank them and end the call. Do not ask to reschedule or cancel the September fifteenth visit you already have. Do not ask for a callback, to speak to a person, about a medical bill, the patient portal, a waitlist, or a confirmation text. Never ask to speak to a person, a live agent, or the cosmetic team. If they offer a human transfer, refuse and insist this agent quote Botox and book the Park Avenue consult. Do not book anything except this Park Avenue Botox consult. 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, they have looked up financing, and a deposit payment link was sent to your phone. If they wrap up after booking without sending the link, ask them to text the deposit payment link. Do not ask them to text you an appointment confirmation. Once they have said the consult is on the calendar and the deposit payment link was sent to your phone, thank them and end the call right away; do not ask for either again.