Eight cards, one chassis
Eight NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs in a single 6.5U rackmount: 768 GB of GDDR7 across PCIe Gen5, enough headroom to serve 70B-class models to 100–500 concurrent users without leaving the rack.
EMARQUE built
EMARQUE AI Server
8× NVIDIA RTX PRO 6000 Blackwell Server Edition (96 GB GDDR7 each, 768 GB total) in a 6.5U rackmount: the EMARQUE AI PRO 800. Supplied and supported through EMARQUE in Malaysia.
768GB GDDR7
8RTX PRO 6000 SE GPUs
8 TBECC DDR5 max
Key features
Configuration overview.
Manufacturer-defined features from the published datasheet.
Datacentre-grade chassis, EMARQUE delivered
In Win IW-RG658-PRO 6.5U platform (a 4.5U GPU chamber plus 2U system chamber, purpose-built for the 8× RTX PRO 6000 SE thermal envelope), supplied through EMARQUE with local commissioning, warranty handling, and Tier-1 support in Malaysia.
Built to run multi-tenant
MIG-style partitioning per card lets you carve the server into isolated inference workloads: separate teams, separate models, separate quotas on shared hardware. No noisy-neighbour fights, full observability via the BMC.
Storage and network that match the GPUs
8× E1.S NVMe plus 4× 2.5-inch hot-swap bays keep the eight GPUs fed; NVIDIA ConnectX-class SuperNIC networking up to 400 / 800 Gb/s with a 10 GbE management port for scale-out clusters. No PCIe contention, no storage starvation under sustained load.
Operates inside Malaysian DC reality
Redundant 3+1 CRPS Titanium power (9,600 W) with 14 hot-swap fans and front-to-back airflow, engineered to run reliably in a Malaysian rack or data-centre environment. EMARQUE confirms site power and cooling readiness during scoping.
Configurable without a redesign
GPU count, memory (up to 8 TB ECC DDR5), storage (E1.S NVMe + 2.5-inch bays), networking (ConnectX SuperNIC 400 / 800 Gb/s), and CPU choice (Intel Xeon or AMD EPYC). All selectable at quote. Need six GPUs instead of eight? The EMARQUE AI PRO 700 is the 6-GPU tier.
Architecture
Under the hood.
The four sub-systems that determine real-workload behaviour. We tune each before delivery.
GPU complex
- Up to 8 × NVIDIA RTX PRO 6000 Blackwell Server Edition (96 GB GDDR7 each)
- 24,064 CUDA cores · 752 5th-gen Tensor cores · 188 4th-gen RT cores per GPU
- FP4 / FP6 / FP8 inference acceleration (5th-gen Tensor)
- 16 × PCIe Gen5 slots via MCIO: no NVLink on this generation of RTX PRO
CPU + memory
- Dual 5th-Gen Intel Xeon Scalable (up to 350 W each)
- Alternative: Dual AMD EPYC (Zen 5), selectable per workload
- High PCIe Gen5 lane count across both sockets
- Up to 8 TB ECC DDR5 across 32 DIMM slots (RDIMM, up to 4800 MHz)
Storage & networking
- 8 × E1.S NVMe hot-swap + 4 × 2.5-inch hot-swap bays (PCIe Gen5)
- RAID 0/1/10 across NVMe
- NVIDIA ConnectX-class SuperNIC up to 400 / 800 Gb/s · 10 GbE management
- InfiniBand available for cluster scale-out
Power, cooling, management
- 9,600 W: 3+1 redundant CRPS, 80+ Titanium
- 14 hot-swap PWM fans (10 GPU + 4 system), front-to-back airflow
- BMC with Redfish API, iKVM, signed firmware updates
- EMARQUE assembly + multi-point QA in Malaysia
Next step
Tell us your workload. EMARQUE sizes the AI Server and sends a quote.
Supported workloads
Reference workload categories.
Workload categories documented in the manufacturer's reference materials. Sizing is confirmed with your technical team during scoping.
Departmental RAG
Internal knowledge chat for 500 users
Run a private RAG stack with a 70B-class model against your document store, code repos, and ticketing system. Eight GPUs split four ways gives four logical inference endpoints sized for 100+ concurrent users each: the right shape for finance, legal, engineering, and ops to share one server.
Production inference
24/7 multi-tenant model serving
TGI / vLLM / Triton serving multiple fine-tuned variants of a base model. PCIe Gen5 isolation per card means no NVLink coherence overhead: the right architecture when each request fits on one GPU and you want predictable per-tenant throughput rather than tightly coupled training.
Vision + voice
Real-time multi-stream processing
Eight independent cards parallelise across camera or audio streams cleanly: object detection on 50 4K streams, speech-to-text on 200 concurrent calls, or pose estimation across a factory floor. Each stream gets a dedicated GPU slice with consistent latency.
Mixed workload
Inference today, fine-tune at night
Run production inference on six cards during business hours, reallocate to LoRA / QLoRA fine-tuning runs on all eight overnight. The BMC + Redfish API makes the rebalance scriptable; no separate dev cluster required.
Full spec sheet
Every line documented at quotation.
Configurable. Final BOM, GPU mix, RAM and storage, and networking topology are confirmed in writing at quotation.
- Model
- EMARQUE AI PRO 800
- GPU
- Up to 8 × NVIDIA RTX PRO 6000 Blackwell Server Edition · 96 GB GDDR7 with ECC (768 GB total)
- AI compute
- 32 PFLOPS FP4 aggregate · 192,512 CUDA cores across 8 GPUs
- GPU memory bandwidth
- 1.6 TB/s per card
- GPU interconnect
- 16 × PCIe Gen5 via MCIO (no NVLink, independent cards)
- CPU
- Dual 5th-Gen Intel Xeon Scalable (up to 350 W each) or AMD EPYC, selectable
- Memory
- Up to 8 TB ECC DDR5 (32 DIMMs, up to 4800 MHz)
- Storage
- 8 × E1.S NVMe hot-swap + 4 × 2.5" hot-swap bays
- Networking
- NVIDIA ConnectX-class SuperNIC up to 400 / 800 Gb/s · 10 GbE management
- Power
- 9,600 W: 3+1 redundant CRPS, 80+ Titanium
- Cooling
- 14 hot-swap PWM fans (10 GPU + 4 system), front-to-back airflow
- Form factor
- In Win IW-RG658-PRO 6.5U rackmount (4.5U GPU + 2U system), 19" rails
- Software
- Ubuntu LTS + NVIDIA AI Enterprise stack
- Warranty
- 3-year EMARQUE warranty · Malaysia-based local support
FAQ
Common questions about AI Server
What is the NVIDIA RTX PRO 6000 Blackwell Server Edition?
Server-optimised variant of the RTX PRO 6000 Blackwell: same Blackwell silicon and 96 GB GDDR7 memory, with passive cooling (300 W TDP) designed to be cooled by the server chassis airflow. Targets enterprise rackmount deployments where the active 600 W variant would be impractical.
What reference platform does EMARQUE use?
EMARQUE supplies the AI Server, the EMARQUE AI PRO 800, on the In Win IW-RG658-PRO 6.5U rackmount platform: a 4.5U GPU chamber plus a 2U system chamber, purpose-built for the 8× RTX PRO 6000 SE thermal envelope. Final BOM is documented at quotation.
Is this configuration customisable?
Yes. GPU count (the 6-GPU tier is the EMARQUE AI PRO 700), memory (up to 8 TB ECC DDR5), storage (E1.S NVMe + 2.5-inch bays), networking (ConnectX SuperNIC 400 / 800 Gb/s + InfiniBand), and OS are all configurable. CPU choice (Intel Xeon vs AMD EPYC) is customer-selectable per workload.
What's the warranty?
A 3-year EMARQUE warranty with Malaysia-based local support on the AI PRO 800, plus NVIDIA's warranty entitlement on the RTX PRO 6000 SE GPUs. EMARQUE handles warranty and RMA locally.
What support does EMARQUE provide locally?
EMARQUE handles in-country delivery, customs, commissioning, acceptance testing, and Tier-1 support response in Malaysia. Tier-2/3 escalation routes to the manufacturer per the warranty entitlement. Optional service contracts for extended response SLAs and on-site engineering visits can be added.
If I need NVLink-coherent multi-GPU or HGX-class, what's the upgrade path?
Step into NVIDIA DGX B200 or NVIDIA DGX B300 (or HGX B200 / B300 OEM platforms from Dell, Giga Computing, or Supermicro). Those configurations are presented on the respective NVIDIA DGX product pages. Each shows both NVIDIA-branded DGX systems and HGX OEM alternatives.
Request configuration & quotation.
Manufacturer specifications and warranty terms apply. EMARQUE issues a formal quotation through your Key Account Manager.
02Talk to EMARQUE
Tell us about your workload.
Model size, concurrency, latency budget, deployment site. EMARQUE returns a quote in MYR within one Malaysian business day, sized to the workload, not the salesperson’s quota.
- 01
Key Account Manager
+6012 627 2280 - 02
Request for Quotation
business@emarque.co
Enquire about the EMARQUE AI Server
Enterprise AI servers are supplied for deployment in Malaysia only, and are subject to end-user, application and location verification. Start with the short pre-screening form.