AI workstations and servers.
Desk
Desk-side systems for one person or a small team: the NVIDIA DGX Spark, delivered and set up by us, and the AI Work 100, assembled and tested by us in Malaysia.

NVIDIA DGX Spark
A desk-side AI computer on NVIDIA's GB10 Grace Blackwell Superchip, with 128 GB of unified memory. For developers and small teams running models on their own machine.
- Unified memory
- 128 GB
- FP4 with sparsity
- 1 PFLOP
- Parameters on one unit
- 200B

AI Work 100
A tower AI workstation with one or two NVIDIA RTX PRO 4500, 5000 or 6000 Blackwell GPUs, for a developer or a small team running models on their own hardware. Assembled and tested by us in Malaysia.
- Max GPU memory per card
- 96 GB
- RTX PRO Blackwell GPUs
- Up to 2
- Max system memory
- 192 GB
Workstation
Bigger machines for a team: the NVIDIA DGX Station beside a desk, and our four-GPU AI PRO 500 in a 5U rackmount chassis.

NVIDIA DGX Station
A desk-side system on NVIDIA's GB300 Grace Blackwell Ultra Desktop Superchip, with 748 GB of coherent memory. For research and data science teams running large models on their own hardware.
- Coherent memory
- 748 GB
- FP4 with sparsity
- 20 PFLOPS
- ConnectX-8 networking
- 800 Gb/s

AI PRO 500
Up to four NVIDIA RTX PRO 6000 Blackwell GPUs on an AMD Threadripper PRO workstation platform, in a 5U rackmount chassis. Built and supported by us in Malaysia.
- GPU memory with four cards
- 384 GB
- RTX PRO 6000 GPUs
- Up to 4
- Rackmount
- 5U
Server
Rackmount servers and NVIDIA's rack-scale systems, grouped by how they are cooled.
Air-cooled
For a server room with standard airflow: our six- or eight-GPU AI Server and AI HGX B300, and the NVIDIA DGX B200 and DGX B300.

EMARQUE AI Server
An AI server with six or eight NVIDIA RTX PRO 6000 Blackwell GPUs, and 576 GB or 768 GB of GPU memory, in one 6.5U rackmount. Built and supported by us in Malaysia.
- GPU memory
- 576 / 768 GB
- RTX PRO 6000 GPUs
- 6 or 8
- Rackmount, air-cooled
- 6.5U

EMARQUE AI HGX B300
Eight NVIDIA Blackwell Ultra GPUs on one HGX B300 board, with 2.1 TB of GPU memory between them. Built to order for training large models, or serving one to a whole organisation.
- GPU memory
- 2.1 TB
- Blackwell Ultra GPUs
- 8
- Max system memory
- 4 TB

NVIDIA DGX B200
Eight NVIDIA Blackwell GPUs linked by NVLink in one air-cooled 10U system, with 1,440 GB of HBM3e. For training and inference on your own premises.
- Blackwell GPUs
- 8
- HBM3e
- 1,440 GB
- FP4 with sparsity (NVIDIA figure)
- 144 PFLOPS

NVIDIA DGX B300
Eight NVIDIA Blackwell Ultra GPUs in one air-cooled 10U system, with 2.1 TB of HBM3e. For reasoning models and inference at scale.
- Blackwell Ultra GPUs
- 8
- HBM3e
- 2.1 TB
- FP4 with sparsity
- 144 PFLOPS
Liquid-cooled
For rooms with coolant distribution: our AI HGX Rubin, and NVIDIA's GB300 NVL72 and Vera Rubin NVL72 racks, which we source through NVIDIA's OEM partners.

EMARQUE AI HGX Rubin
Eight NVIDIA Rubin GPUs on one HGX Rubin NVL8 board, with 2 TB of GPU memory between them, liquid-cooled. Built to order for training large models, or serving them to a whole organisation.
- HBM4 GPU memory
- 2 TB
- Rubin GPUs
- 8
- NVFP4 inference, sparse
- 400 PF

NVIDIA GB300 NVL72
Thirty-six NVIDIA Grace CPUs and 72 NVIDIA Blackwell Ultra GPUs in one liquid-cooled rack, linked by NVLink. We source it through NVIDIA's OEM partners and coordinate the install.
- Blackwell Ultra GPUs
- 72
- GPU memory
- 20 TB
- NVLink bandwidth
- 130 TB/s

NVIDIA Vera Rubin NVL72
NVIDIA's liquid-cooled rack-scale platform on the Rubin generation, the step after NVIDIA GB300 NVL72. We source it through NVIDIA's OEM partners and coordinate the install.
Specifications on enquiry
Learn moreTell us about your workload.
Tell us what it needs to do and who will use it. A few lines is enough. We reply by email, and every quote comes in writing, in MYR.
