Skip to content
EMARQUE AI

Compare

Compare our systems.

Every system we offer, grouped by class, with the figures buyers compare first. Quoted on request.
Desk

2 systems in this class, side-by-side.

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.

DGX Spark

Best for
Developers and small teams running models on their own machine
Form factor
Compact desktop
GPU memory
128 GB unified
Configurable?
NVIDIA configuration, choice of OEM variant

AI Work 100

Best for
One developer or a small team running open-weight models on their own hardware
Form factor
Tower workstation, standard 240 V wall socket
GPU memory
32 to 192 GB (1 or 2 × 32, 48 or 96 GB)
Configurable?
Yes, built to order
Workstation

2 systems in this class, side-by-side.

Bigger machines for a team: the NVIDIA DGX Station beside a desk, and our four-GPU AI PRO 500 in a 5U rackmount chassis.

DGX Station

Best for
Running models up to about a trillion parameters on one machine (NVIDIA figure)
Form factor
Desk-side tower
GPU memory
748 GB coherent (252 GB HBM3e)
Configurable?
NVIDIA configuration, choice of OEM variant

AI PRO 500

Best for
A private chat and document-search service for one department
Form factor
SilverStone RM52 5U rackmount, 19-inch rack
GPU memory
384 GB (4 × 96 GB)
Configurable?
Yes, built to order
Server, air-cooled

4 systems in this class, side-by-side.

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.

AI Server

Best for
A private chat and document-search service shared by several departments
Form factor
In Win 6.5U rackmount (4.5U GPU + 2U system): IW-RG658-PRO for eight GPUs, IW-RG650A for six
GPU memory
576 or 768 GB (6 or 8 × 96 GB)
Configurable?
Yes, built to order

AI HGX B300

Best for
Training and fine-tuning models too large for one GPU
Form factor
8U, 19-inch rackmount
GPU memory
2.1 TB HBM3e
Configurable?
Yes, built to order

DGX B200

Best for
Training and fine-tuning large models inside one node
Form factor
10U rackmount · 444 × 482 × 897 mm · up to 142.4 kg
GPU memory
1,440 GB HBM3e
Configurable?
NVIDIA configuration, or an OEM build of the same platform

DGX B300

Best for
Reasoning models and long-context inference
Form factor
10U rackmount, air-cooled
GPU memory
2.1 TB HBM3e
Configurable?
NVIDIA configuration, or an OEM build of the same platform
Server, liquid-cooled

3 systems in this class, side-by-side.

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.

AI HGX Rubin

Best for
Training and fine-tuning large models on NVIDIA's Rubin generation
Form factor
On enquiry
GPU memory
2 TB HBM4
Configurable?
Yes, built to order

GB300 NVL72

Best for
Training models too large to fit in one server
Form factor
One liquid-cooled rack
GPU memory
20 TB HBM3e
Configurable?
NVIDIA configuration, or an OEM build of the same platform

Vera Rubin NVL72

Best for
Rack-scale deployments on NVIDIA's Rubin generation
Form factor
One liquid-cooled rack
GPU memory
On enquiry
Configurable?
NVIDIA configuration, or an OEM build of the same platform

Tell 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.