88 threads and two independent 72 GB GPUs mean this single tower can genuinely run two
workstreams at once — not just a bigger version of a one-job-at-a-time machine.
144 GB of aggregate VRAM, split however the work needs it.
144 GB Aggregate VRAM44c / 88t CPUUbuntu ReadyAssembled in USA · TAA
2x 72 GB
GDDR7 GPUs, 144 GB aggregate
44c / 88t
Intel Xeon W9-3575X
128 GB
DDR5, ECC Registered
6 TB
NVMe SSD storage installed
2000 W
80 PLUS Gold PSU
AI Silicon Spotlight · Hardware First
What This Pairing Actually Buys You
The highest CPU thread count in this chassis lineup's dual-GPU tier, feeding two independent 72GB GPUs:
GPU · 2x NVIDIA RTX PRO 5000 Blackwell
144 GB Aggregate, Two Independent Cards
Two independent 72 GB GPUs — 144 GB aggregate, not a single unified pool. Run two workloads, or split one large job across both.
Blackwell 5th-gen Tensor Cores with FP41 accelerate agentic and generative AI on each card.
4th-gen RT Cores1 drive photoreal rendering previews on either GPU.
Universal MIG1 can partition either card further for multiple concurrent users.
CPU · Intel Xeon W9-3575X
44 Cores, 88 Threads, Two GPUs to Feed
44 cores / 88 threads on LGA4677 — the highest thread count Intel option in this chassis lineup's dual-GPU tier.
Notably more cores and threads than this lineup's AMD-based dual-GPU alternative, on the same Compare Options list — a genuine platform difference, not a universal rule about either vendor.
8-channel DDR5 memory keeps both GPUs supplied with data without the CPU becoming the shared bottleneck.
Workstation-grade ECC memory support brings server-class reliability to long dual-GPU jobs.
Two GPUs, Two Workstreams
One Box, Many Jobs at Once
88 threads and two independent GPUs turn "one job at a time" into "two jobs, right now." Some concrete pairings this configuration can actually run:
GPU 1 Runs
Fine-tune a model against a training set
Render a production scene
Encode a finished video export
Run the game engine live
GPU 2 Runs, at the Same Time
Serve live inference requests
Preview the next scene's changes
Keep editing the next timeline
Test the previous build in parallel
Illustrative workload pairings, not a guaranteed performance benchmark for any specific software.
AI Features · What This Actually Does
Six Jobs, Two GPUs Deep
The listing's own AI Features, translated into what running two GPUs and 88 threads actually enables:
AI Development
Fine-tune on one GPU while serving inference on the other, with 88 threads keeping both fed with data.
Data Science
88 threads handle preprocessing for two parallel data pipelines, one feeding each GPU.
HPC
Split a large simulation across both GPUs, or run two independent simulations side by side.
AI-Driven Rendering & Graphics
Render two scenes at once on separate GPUs, or split a single large scene's passes across both.
Video Content & Streaming
Encode one project while editing the next, one GPU each, with 88 threads handling the pipeline work around both.
Game Development
Run the engine live on one GPU while a build tests on the other — one box, two workstreams.
Scene photos are AI-generated illustrations of typical deployments, not photographs of this exact product.
OpenClaw agent framework (pre-installed); Qwen 3.5 27B+ local LLM package
FAQ
Quick Answers Before You Buy
Is the 144 GB of VRAM one big pool or two separate GPUs?
Two independent 72 GB GPUs — 144 GB aggregate, not a single unified pool. Run two separate workloads, one per GPU, or split a single large job across both.
Why 44 cores / 88 threads specifically?
Two GPUs need enough CPU parallelism to keep both fed with data at once. This is the highest core/thread count Intel option in this chassis lineup's dual-GPU 144GB tier — notably more than the AMD-based dual-GPU option in the same tier.
Can this really run two separate jobs at the same time?
Yes — with two independent GPUs and 88 CPU threads, this is a genuinely different capability from a single-GPU workstation. Train on one GPU while serving inference on the other, for example.
Does it come with an AI agent framework or local LLM pre-installed?
The AI agent framework (OpenClaw) is pre-installed; a Qwen 3.5 27B or higher local LLM package is an optional add-on configured at order time.
Is it eligible for government / public-sector purchasing?
It is TAA compliant and assembled in the USA, with compliance documents available for purchase review.
How is it supported?
1-year parts and labor limited warranty backed by ABS manufacturer support. As a Build to Order system, it is custom built and stress-tested after purchase.
Assembled in USA
TAA Compliant
144 GB Aggregate VRAM
Build to Order flexibility
1. Intel-published core/thread count for the Xeon W9-3575X. GPU architecture features (5th-gen Tensor Cores, 4th-gen RT Cores, Universal MIG) are NVIDIA-published for the Blackwell professional GPU architecture generation, not SKU-specific benchmark figures.
ABS Zaurion Ruby ZAW9-3575X-RP500072X2 · Gigabyte W773-W80 platform (LGA4677 socket) · Optional components and software are configured at time of purchase.
Learn more about the ABS ZAW9-3575X-RP500072X2
Model
Brand
ABS
Series
Zaurion Ruby
Model
ZAW9-3575X-RP500072X2
Part Number
ZAW9-3575X-RP500072X2
Details
Workload
AI Workstation
Type
Tower
AI Features
AI Development: Accelerate AI development and inference workloads, and build agentic AI applications. Data Science: Accelerate your data pipelines. HPC: Accelerate scientific breakthroughs with enhanced accuracy. AI-Driven Rendering and Graphics: Deliver stunningly realistic designs and simulations with speed and precision. Video Content and Streaming: Boost performance for live media and video pipelines. Game Development: Drive new possibilities for game development with AI.
Processor
CPU Type
Intel Xeon W9-3575X
CPU Socket
LGA 4677
Number of CPU Cores
44
# of CPU Threads
88
# of Processors Installed
1
MAX Processors
1
Chipset
Intel W790
Memory
Installed Memory Size
128GB
Memory Type
DDR5
Memory Features
ECC
Number of Memory Slots
8
Storage
SSD (Installed)
OS SSD: 2TB M.2 Data SSD: 4TB M.2
Graphic Controller
Servers_Graphic Installed
2x NVIDIA RTX Pro 5000 Blackwell 72GB
GPU Type
NVIDIA RTX Pro 5000 Blackwell
Video Memory
144GB
Networking
LAN Ports
2 x 10 Gb/s LAN 1 x 10/100/1000 Mb/s Management LAN
Software
Operating System
Ubuntu
Power
Power Characteristics
Single 2000W ATX 80 PLUS Gold power supply
Energy-Efficient
2000W
Features
Features
GIGABYTE W773-W80 GPU Workstation
Additional Information
First Listed on Newegg
August 20, 2026
Customer Reviews of the ABS ZAW9-3575X-RP500072X2
There are no reviews yet.
Do you own this product?
Warranty & Returns
Warranty, Returns, And Additional Information
Warranty
Limited Warranty period (parts): 1 year
Limited Warranty period (labor): 1 year
Return Policies
Return for refund within: non-refundable
Return for replacement within: non-replaceable
This item is covered by NeweggBusiness.com's Manufacturer Only Return Policy.
