ABS Zaurion Duo Ruby
A Ready-to-Deploy Dual-GPU Workstation Powerhouse That Gives You a Stylish workspace
Pioneer new frontiers in AI, data science, and graphics.
With Dual NVIDIA RTX PRO 6000 Blackwell Max-Q
Perfect for multi-GPU workloads like large-scale data modeling, distributed rendering, or multi-instance AI training and inference.
AI Development
Accelerate AI development and inference workloads, and build agentic AI applications.
Data Science
Accelerate your data pipelines.
AI-Driven Rendering and Graphics
Deliver stunningly realistic designs and simulations with speed and precision.
HPC
Accelerate scientific breakthroughs with enhanced accuracy.
Video Content and Streaming
Boost performance for live media and video pipelines.
Powered by NVIDIA Blackwell Architecture
Fifth-Gen Tensor Cores
Max AI performance with FP4 and DLSS 4
New Streaming Multiprocessors
Optimized for RTX Neural Shaders
Fourth-Gen Ray Tracing Cores
Built for RTX Mega Geometry
5th Gen Tensor Cores
Fifth-generation Tensor Cores deliver up to 3X the performance of the previous generation and add support for FP4 precision and DLSS 4 Multi Frame Generation technology. Accelerate local LLMs, prototype new AI models, and drive enhanced content creation and graphics.
4th Gen RT Cores
Fourth-generation RT Cores deliver up to 2X the performance of the previous generation, accelerating rendering for M&E content creation, AECO design, and manufacturing prototyping. Create photoreal, physically accurate scenes and immersive 3D designs with neural graphics-based technologies, such as RTX Mega Geometry, enabling up to 100X more ray-traced triangles.
CUDA Cores
NVIDIA Blackwell is the most powerful professional RTX GPU ever created with the latest SM and CUDA® core technology. The SM features increased processing throughput, and new neural shaders that integrate neural networks inside of programmable shaders to drive the next decade of AI-augmented graphics innovations.
96GB of GPU Memory
New and improved GDDR7 memory significantly boosts bandwidth and capacity, empowering your applications to run faster, and work with larger, more complex datasets. With 96GB of GPU memory, tackle massive 3D and AI projects, explore large-scale VR environments, and drive larger multi-app workflows.
9th-Gen NVENC
Ninth-generation NVIDIA NVENC engines significantly accelerate video encoding speed, and improve quality for professional video applications. They add new support for 4:2:2 H.264 and HEVC encoding, and improve HEVC and AV1 encoding quality.
6th-Gen NVDEC
Sixth-generation NVIDIA NVDEC engines provide up to double H.264 decoding throughput and offer support for 4:2:2 H.264 and HEVC decode. Professionals can benefit from high-quality video playback, accelerate video data ingestion, and use advanced AI-powered video editing features.
Multi-Instance GPU
Multi-Instance GPU (MIG) expands the performance and value of RTX PRO 6000 Max-Q by enabling the creation of up to four (4) fully isolated instances. Each MIG instance has its own high-bandwidth memory, cache, and compute cores. and guaranteed quality of service (QoS), extending the reach of accelerated computing resources to every user.
PCIe Gen 5
Support for PCI Express Gen 5 provides double the bandwidth of PCIe Gen 4, improving data-transfer speeds from CPU memory and unlocking faster performance for data-intensive tasks like AI, data science, and 3D modeling.
Max-Q
Packing a 300W punch into an ultra-efficient design, the RTX PRO 6000 Max-Q redefines high-density workstation performance. Optimized for energy and space efficiency, it enables up to 4 GPUs in a single system, delivering unmatched scalability for AI training, 3D rendering, and complex simulations.
Industry-Leading Performance
AI Performance
3511 TOPS1,2
RT Core Performance
333 TFLOPS1
Single-Precision Performance
110 TFLOPS1
AMD Ryzen™ Threadripper™ PRO 7000 WX-Series Processors
Next-Gen Power
AMD Ryzen™ Threadripper™ PRO 7000 WX-Series processors expand on the prior generation’s unrivaled performance and superior platform features. Built on the leading 5nm ‘Zen 4’ architecture, these processors deliver power, performance, expandability, and efficiency to your fingertips. Common lightly threaded applications used by artists, architects, and engineers benefit from enhanced max boost frequencies and IPC improvement across the entire stack, delivering outstanding performance.
For multithreaded workloads, Threadripper™ PRO continues to offer the most cores of any workstation processor for complex simulation, generative design, rendering, and software compilation tasks where users can see significant performance gains in applications. Additionally, these processors offer up to 1.5 times the L3 cache directly accessible per core compared to the previous generation for an effective reduction in memory latency, along with high memory bandwidth for applications that are sensitive to memory speed.
Zen 4
architecture
AMD WRX90
chipset
Up to 96
CPU cores
8-channel
DDR5 memory
Up to 128
PCIe 5.0 lanes
AST2600
BMC
Increase Productivity
Save up to 27 minutes for every hour of productivity on the fastest workstation processors on the market.3
Designed for artists, architects, and engineers, these processors excel in both multi-threaded and lightly threaded workflows, offering exceptional power, performance, and platform expandability built on the leading Zen 4 architecture.
Create Without Restraint
Fast-track creative workflows with the performance, value, and reliability of next-gen professional GPUs and high-speed storage devices.
AMD Ryzen Threadripper processors support a range of advanced workstation configurations to meet the needs of cutting-edge visual work.
Accelerate AI Workflows
Enhance your engineering, creative, scientific and development workfloads and avoid paying expensive cloud subscription costs, with on-device genAI like LLM and diffusion models.
From conception to final production, integrate advanced AI tools into every phase of your project.
Professional Thermal Solution
Internal airflow diversion ensures performance and heat dissipation
Storage Support
Mix Drives Support
Plentiful NVMe and SATA Storage interface can fulfill your hot and cold data processing requirement at once.
- Onboard 4 x M.2 2280/22110 slots (PCIe 5.0 x4)
- 4 x 3.5” / 2.5” SATA hot-swap HDD/SSD bays
- Optional 4 x 3.5” / 2.5” SATA hot-swap HDD/SSD bays
Connectivity
Dual 10GbE LAN
The Broadcom® BCM57416 controller supports 2 x 10GbE configuration which is the foundation for workstation connectivity, providing broad interoperability, critical performance optimizations, and increased agility for Communications, Cloud, and Enterprise IT network solutions.
Optional AI Memory Optimization with Phison aiDAPTIV+
This system supports optional Phison aiDAPTIV+ SSD to significantly reduce GPU VRAM requirements for Full Fine-Tuning of large language models.
Key Message
With optional aiDAPTIV+, customers can fine-tune large LLMs using much smaller GPUs, dramatically lowering hardware cost and deployment complexity.
| Target Model Size | VRAM Needed (Without aiDAPTIV+) | VRAM Needed (With aiDAPTIV+) | Typical Hardware |
|---|---|---|---|
| = 13B | 120-200 GB | 32 GB | 1x RTX 5090 |
| = 34B | 250-350 GB | 96 GB | 1x RTX PRO 6000 |
| = 70B | 500-700GB | 192 GB | 2x RTX PRO 6000 |
What this means
- Without aiDAPTIV+, Full Fine-Tuning typically requires multi-GPU servers
- With optional aiDAPTIV+, the same model can be trained on a single workstation
- This enables lower cost, simpler deployment, and faster AI adoption
Phison aiDAPTIV+ SSD is optional add-on and is not included by default.
Product Overview
1. Peak rates based on GPU Boost Clock.
2. Theoretical FP4 TOPS using sparsity.
3. Based on AMD performance lab testing as of July, 2023, using the SPEC Workstation 3.1 Media and Entertainment metric, SPEC Workstation 3.1 Product Development PTC Creo subset metric, SPEC Workstation 3.1 Life Sciences overall metric, SPEC Workstation 3.1 Financial Services overall metric, SPEC Workstation 3.1 Energy overall metric, the V-Ray CPU performance benchmark, the PugetBench for Adobe AfterEffects v0.95.7 benchmark, the PugetBench for Premiere Pro v0.98.0 benchmark, the SPECapc Maya 2023 CPU composite metric, the Unreal Engine 5.1 compilation benchmark, the Chromium Compilation 115.0.5740 benchmark, the SPECapc for Solidworks 2022 CPU composite metric, the Keyshot CPU rendering benchmark, the Puget Metashape total Processing time (Rock Model) benchmark, and the Corona rendering (Rays/Sec) benchmark, Blender Rendering Benchmark, to compare the performance of AMD Ryzen Threadripper PRO 5995WX, 5975WX, 5965WX, 5955WX, 5945WX processors in reference systems configured with 8x64GB DDR4, NVIDIA Quadro RTX A5000, 1TB SSD, Win 11 vs. a similarly configured BOXX workstation with Intel Xeon w9-3495X, w9-3475X, w7-3455X, w5-3435X, w5-3425X, w7-2495X, w7-2465X processors Workstation manufacturers may vary configurations, yielding different results. Results may vary. CGP-45