Dual RTX PRO 6000 Blackwell LLM Workstation (192GB VRAM, 1TB ECC)
Two Blackwell GPUs and 192GB VRAM for serving and fine-tuning the largest local models.
We help you choose, configure, and deliver the right system — no obligation.
Configuration at a Glance
Tailored per engagement. Full technical overview below.
Configuration Options
Core specifications for this system. Every component is configurable to your workload — request a quote for a tailored build.
2× NVIDIA RTX PRO 6000 Blackwell — 192GB combined GDDR7 ECC
AMD Threadripper PRO 7000-series WX (64-core class)
Up to 1TB DDR5 ECC
8TB PCIe Gen5 NVMe for model weights
Overview
This flagship workstation combines two NVIDIA RTX PRO 6000 Blackwell GPUs for 192GB of combined ECC VRAM, backed by up to 1TB of system ECC memory, for teams building and serving frontier-scale language models on-premises. Nexus Compute sources and validates the full dual-GPU configuration through authorized channels, tuning thermals and the inference stack so it operates as a stable production platform.
Who This Solution Is For
Business Benefits
Frontier models stay local
192GB of combined ECC VRAM runs models and context lengths that otherwise force teams onto rented cloud GPUs.
Predictable economics
A fixed capital purchase replaces unpredictable per-hour and per-token cloud costs for sustained development.
Validated end-to-end
We test the dual-GPU thermal and software stack so throttling and driver issues are resolved before delivery.
Typical Business Use Cases
Local serving of large LLMs with vLLM or TGI
Full fine-tuning and QLoRA on large models
Long-context agentic system development
Parallel multi-model experimentation
Industry Applications
Technical Overview
Two NVIDIA RTX PRO 6000 Blackwell GPUs deliver 192GB of combined GDDR7 ECC across PCIe Gen5 on an AMD Threadripper PRO 7000-series WX platform supporting up to 1TB of DDR5 ECC. The system ships with vLLM and the Hugging Face stack pre-configured on Ubuntu LTS, with high-wattage redundant-grade power and validated cooling.
| GPU / Accelerator | 2× NVIDIA RTX PRO 6000 Blackwell — 192GB combined GDDR7 ECC |
| GPU Interconnect | PCIe Gen5 x16 per GPU |
| CPU | AMD Threadripper PRO 7000-series WX (64-core class) |
| Memory | Up to 1TB DDR5 ECC |
| Storage | 8TB PCIe Gen5 NVMe for model weights |
| Software | vLLM, Hugging Face, CUDA pre-configured |
| Form Factor | Full-tower (air-cooled, validated thermals) |
| Power | 2000W 80+ Titanium |
| Warranty | 3-year on-site NBD + professional GPU warranty |
Specifications are indicative and configured to each engagement. Request a quote for a configuration tailored to your requirements.
Warranty, Support & Fulfillment
Every system ships from an authorized channel, configured and tested, with the documentation enterprise buyers need — backed by warranty and a dedicated account team.
Enterprise Warranty
Full manufacturer warranty with optional on-site, next-business-day support and extended coverage.
Authorized Channel
Sourced through Tier-1 distribution and OEM partners — never grey market. Asset & warranty records included.
Lead Time & Deployment
48-hour quotes, then configured, burn-in tested, and delivered on a committed schedule.
Nationwide Fulfillment
Coordinated logistics, rack-and-stack, and delivery wherever your infrastructure lives.
Frequently Asked Questions
Do the two GPUs use NVLink?
RTX PRO 6000 Blackwell workstation cards communicate over PCIe Gen5 rather than NVLink; modern inference and training frameworks shard across PCIe effectively, and 192GB of pooled VRAM is the practical benefit. We tune the framework configuration for your model sizes.
How large a model can this run?
192GB of VRAM comfortably serves large open-weight models in full or near-full precision and fine-tunes them with QLoRA. We size quantization and batch settings to your specific models during configuration.
When should I move to a GPU server instead?
Move to a rackmount server when you must serve many concurrent production users or train beyond a single node. We can advise on our quad-GPU and GPU Server lines as the next step.
Hardware Assistance
Configure the Dual RTX PRO 6000 Blackwell LLM Workstation (192GB VRAM, 1TB ECC) with Nexus Compute
Tell us your requirements and a hardware specialist will help you specify, configure, and quote the right system — typically within two business days. No obligation.