The conversational AI stack, running on hardware inside your building.
No cloud dependency, no per-minute fee, no conversation data leaving the building. Configurations sized by call load, built to spec.
Desk units
Standard outlet, no rack, no server room. Sits in the room it serves.
B-01
One point of contact. Supports up to 5 concurrent sessions.
- FORM SFF TOWER · ~15L
- VRAM 32GB
- PEAK DRAW 350W
- HEAT 1,194 BTU/H
- PSU 650W SINGLE
- NETWORK 1GbE
STARTS AT $4,000
B-02
One site, no rack. Supports up to 10 concurrent sessions.
- FORM MID TOWER · ~30L
- VRAM 64GB
- PEAK DRAW 580W
- HEAT 1,979 BTU/H
- PSU 750W SINGLE
- NETWORK 1GbE
STARTS AT $6,000
Rack units
Redundant power, IPMI, ECC memory, SIP terminating on the unit.
S-01
One site, one unit. Supports up to 15 concurrent sessions.
- FORM 2U RACK · 600MM
- VRAM 80GB
- PEAK DRAW 675W
- HEAT 2,303 BTU/H
- PSU 800W CRPS REDUNDANT
- NETWORK 2× 1GbE + IPMI
STARTS AT $8,000
S-02
Many sites, one campus. Supports up to 30 concurrent sessions.
- FORM 4U RACK · 750MM
- VRAM 128GB
- PEAK DRAW 1,220W
- HEAT 4,163 BTU/H
- PSU 1,600W CRPS REDUNDANT
- NETWORK 2× 1GbE + IPMI
STARTS AT $12,000
The configurations above are preliminary references issued for evaluation. Components, specifications, and prices are indicative and subject to change without notice, and depend on hardware availability at the time of order. Stated capacity figures are design targets and vary with workload and deployment. Units are supplied under a pilot program and are not offered for general sale.