a. Single-GB10 DGX Spark
Cost-optimized entry appliance for focused SME workflows, departmental pilots, and moderate concurrency.
AI Appliance
One Full-Stack AI Harness. Three deployable form factors.
Same vertically optimized stack — from a single GB10 appliance to siloed private cloud.
One Harness · three deployable form factors
Key diagram
Ivertiq Full-Stack Platform
The appliance is the physical embodiment of the stack
Customers buy a working vertical AI system — localized models, governed agents, and workflow apps — not a GPU server plus a model download.
Vertical Apps
Industry workflows, agents, and solutions
AI Harness Control
Orchestration, agents, policies, guardrails, HITL & audit
Ivertiq Core Foundation
Local & vertical LMs · knowledge engine · inference optimization
What changes across SKUs
Capacity
Concurrency & context
More GPUs and siloed cloud capacity for heavier orchestration and more users.
Model class
Best-fit local LMs
Entry tiers for strong local LMs; dual-GB10 for near-frontier configurations — validated per workload during design-partner scoping.
Ops model
Owned vs managed
Customer-owned appliance or multi-tenant private cloud with isolation and auditability.
b. Dual-GB10 DGX Spark
Higher-throughput appliance for heavier agent orchestration, larger context, more users, and near-frontier configs.
c. Private Cloud (siloed)
Multi-tenant, siloed managed option: SaaS-like ops with tenant isolation, controlled serving, and auditability — without public LLM API dependency.