HP Inc BM6P1UT#ABA Z2 Mini G1i SmartBuy Workstation Desktop
Overview
The HP Z2 Mini G1i is a compact workstation desktop that delivers 15th-generation Intel Core Ultra processing and professional-grade NVIDIA Quadro RTX graphics in a form factor small enough to mount behind a monitor or tuck into a server rack shelf — without compromising the compute headroom that workstation deployments demand. The BM6P1UT#ABA ships as a SmartBuy configuration: a pre-validated, channel-stocked build that eliminates custom-order lead times while locking in a proven hardware combination for volume deployments.
This unit is built around Intel's 14-core Core Ultra 5 235 (2.90 GHz base, up to 5.00 GHz boost) paired with Intel AI Boost at up to 13 TOPS — the neural processing unit built directly into the SoC. Combined with the discrete NVIDIA Quadro RTX A4-series GPU, this machine handles GPU-accelerated analytics workloads, VMS server-side decoding, AI inferencing at the edge, and CAD or media-review workflows that would overwhelm a business-class mini PC. Explore the full HP Inc workstation and compute lineup for platform context.
Key Features
- Intel Core Ultra 5 235 — 14 Cores, Up to 5.00 GHz: The hybrid core architecture (performance + efficiency cores) distributes compute load intelligently: high-frequency tasks like real-time video decoding spike to 5.00 GHz on P-cores while background indexing and OS overhead runs on E-cores without burning the power budget. This matters on a VMS host running 32+ camera streams simultaneously.
- Intel AI Boost NPU — Up to 13 TOPS: The dedicated neural processing unit offloads AI inferencing from the CPU and GPU, keeping both free for other tasks. At 13 TOPS, it can handle on-device object classification, anomaly detection, and metadata generation without the latency penalty of cloud round-trips — a real factor in air-gapped or low-latency security deployments.
- NVIDIA Quadro RTX A4-Series GPU: Quadro RTX silicon is certified for professional applications (VMS GPU decoding, forensic video review, AI model inference) where consumer GeForce cards lack driver-level ISV validation. For integrators running GPU-accelerated video management software or AI analytics engines, this is the tier that gets you on vendor support matrices.
- 16 GB RAM: A functional baseline for a VMS workstation running a moderate camera count or for a developer/analyst workstation. If your deployment targets 64+ camera channels or runs multiple simultaneous AI pipelines, plan to verify upgrade path to higher memory configurations in the Z2 Mini G1i family before committing at this SKU.
- 512 GB M.2 PCIe NVMe 2280 TLC SSD (4×4 lane): PCIe Gen 4×4 bandwidth keeps OS, application, and index reads from becoming a bottleneck during multi-stream playback or database queries. TLC NAND at this capacity is a sensible OS/application drive; pair it with network-attached or DAS storage for long-retention video archives. The 2280 form factor is the standard full-length M.2, so replacement sourcing is straightforward.
- Intel Graphics (Integrated) + Discrete Quadro RTX: The dual-GPU configuration lets you run display outputs off the integrated Intel Graphics while reserving the Quadro RTX exclusively for compute — useful when the workstation doubles as an analytics node and an operator console.
- Mini Workstation Form Factor: The Z2 Mini G1i's compact chassis is engineered for space-constrained deployments — command center consoles, on-camera-room edge nodes, or back-of-display mounting. At 10.95 lb and 19.00 × 7.00 × 12.00 in, it ships lighter than a tower but carries workstation-class silicon.
- SmartBuy Channel Configuration (BM6P1UT#ABA): SmartBuy SKUs are pre-configured builds sourced through HP's commercial channel, which means faster availability, simplified procurement, and a consistent hardware fingerprint across multi-unit deployments — avoiding the driver-version drift that custom-configured orders can introduce at scale.
Integration & Compatibility
The Z2 Mini G1i's Intel Core Ultra 5 235 platform supports modern NVR and VMS server applications that leverage CPU Quick Sync or GPU-accelerated decoding pipelines. The NVIDIA Quadro RTX A4-series GPU is on ISV-certified hardware lists for major video analytics platforms, AI inferencing frameworks (CUDA-based), and professional visualization tools. The 15th-gen Intel architecture supports PCIe Gen 4, DDR5 memory (verify slot config in HP's full Z2 Mini G1i datasheet for expansion limits), and standard Windows workstation OS images for enterprise imaging and deployment via SCCM or Intune.
For edge AI deployments, the on-die NPU at 13 TOPS can run lightweight OpenVINO or Windows Studio Effects inference workloads without touching CPU or GPU resources — keeping the Quadro RTX available for heavier model inference simultaneously.
Frequently Asked Questions
Q: What GPU is included in the BM6P1UT#ABA?
A: The BM6P1UT#ABA includes an NVIDIA Quadro RTX A4-series discrete GPU alongside Intel integrated graphics on the Core Ultra 5 235 SoC. Verify the exact Quadro RTX sub-model against HP's official QuickSpecs document for this SKU before specifying for GPU-VRAM-sensitive workloads.
Q: What does Intel AI Boost (up to 13 TOPS) actually do in a security or analytics deployment?
A: Intel AI Boost is a dedicated Neural Processing Unit (NPU) built into the Core Ultra 5 235 SoC. It handles AI inference tasks — object detection, scene classification, anomaly flagging — directly on-chip without consuming CPU or GPU cycles, which reduces latency and keeps the rest of the compute stack free for streaming and storage I/O.
Q: Is the 512 GB SSD in the BM6P1UT#ABA sufficient for video storage?
A: The 512 GB NVMe SSD is sized for the OS and applications, not long-retention video archives. Plan on network-attached storage (NAS) or a DAS expansion for video retention. The SSD's PCIe Gen 4×4 interface is well-matched for fast application launch and database queries even under heavy concurrent stream loads.
Q: Can the RAM be upgraded beyond 16 GB?
A: The Z2 Mini G1i platform supports memory expansion, but the exact maximum and available slot configuration should be confirmed against HP's official QuickSpecs for the G1i generation before ordering. The BM6P1UT#ABA ships with 16 GB, which is the SmartBuy baseline.
Q: Is the HP Z2 Mini G1i BM6P1UT#ABA a good fit for a VMS server workload?
A: Yes, for mid-scale VMS deployments. The 14-core Core Ultra 5 235 handles multi-stream decoding, the Quadro RTX GPU accelerates GPU-decode pipelines supported by major VMS platforms, and the NVMe SSD provides fast I/O for concurrent read/write during live recording and playback. For large-scale deployments (100+ cameras), evaluate higher-tier Z2 Tower or rack configurations.
Q: What is the form factor and mounting footprint of the BM6P1UT#ABA?
A: The Z2 Mini G1i measures 19.00 × 7.00 × 12.00 inches and weighs 10.95 lb. HP offers VESA mount adapters for behind-display installation. Confirm mounting accessory compatibility with HP's Z2 Mini G1i accessory catalog for your specific mounting requirement.
The BM6P1UT#ABA is the configuration I'd reach for when a deployment needs a compact analytics-capable node rather than a full tower — the Intel Core Ultra 5 235's NPU at 13 TOPS is what separates this from a standard mini PC, and paired with Quadro RTX, you're holding a machine that can run GPU-accelerated VMS decoding and on-device AI inference simultaneously without either pipeline starving the other.
Technical Highlights:
- 14-Core Intel Core Ultra 5 235 (up to 5.00 GHz): Hybrid core design keeps burst-heavy decode tasks on P-cores at 5.00 GHz while E-cores absorb OS and background workloads — translating to consistent throughput on multi-stream VMS hosts without thermal throttling under sustained load.
- Intel AI Boost NPU — 13 TOPS: Offloads inference from the CPU and GPU entirely. In practice this means you can run a lightweight object-detection model on the NPU while the Quadro RTX handles GPU-accelerated video decode — two compute-heavy tasks running in parallel without resource contention.
- 512 GB PCIe Gen 4×4 NVMe TLC SSD: The 4-lane Gen 4 interface delivers enough sequential bandwidth to handle concurrent application I/O during live recording sessions. TLC at 512 GB is a mature reliability point for an OS/application drive in a 24/7-on workstation context.
Deployment Considerations:
- The 16 GB RAM baseline is adequate for light-to-moderate VMS channel counts, but integrators targeting 64+ streams or multi-model AI pipelines should confirm the Z2 Mini G1i G1 generation's maximum memory support before deploying at volume — the SmartBuy SKU ships as-configured with no field upgrade guarantee documented in the evidence on hand.
- The GPU model designation in the source feed is truncated to 'NVIDIA Quadro RTX A4' — confirm the exact sub-model (and its VRAM spec) via HP QuickSpecs before committing to GPU-VRAM-dependent workloads such as large AI model inference or multi-display professional visualization.
This unit is well-positioned as a space-constrained edge analytics workstation in physical security command centers or operator consoles where rack space is limited — the Mini form factor with workstation-class NPU and Quadro RTX makes it a more defensible choice than a consumer NUC for deployments where GPU driver certification and AI acceleration both matter.
