GLabs Enterprise

Server Families — 1U to 8U

Six rack families, from a single-socket 1U management node to an eight-GPU HGX system. Configured to the workload, validated with GenesysBench™ and deployed through GLIDS, with the GLabs AI Stack added for AI workloads.

By Height

Rendered close view of server racks behind glass doors, with rows of cabled rack servers

Height	Families and Chassis Classes
1U	General Purpose (`GL1U4LW-C262/2L2T` / `GL1U4LW-C252` / `GL1U2S-C262/FIO` / `GL1U2E-C252`) · Storage (all-flash, hybrid and capacity) · Multi-Node (`GL1U2N2G-C252/2T` / `GL1U2N2G-AM5/2T`)
2U	Scale-Out GPU (`GL2U4G-GENOA/M3`) · Storage · Multi-Node (`GL2U4N-F/GENOA-M3`)
4U	Scale-Out GPU (`GL4U8G-` / `GL4U10G-`) · Scale-Up GPU (`GL4U8X-TURIN2/DLC SYN`) · MGX (`GL4UXGM-`) · Storage (`GL4U36L6E-GENOA` / `GL4U60L8E-GENOA`) · Multi-Node (`GL4U18N-`)
6U	Scale-Up GPU (`GL6U8X-EGS2` / `GL6U8X-TURIN2 SYN`) · MGX (`GL6UXGM-GNR2 CX8/DLC`)
8U	Scale-Up GPU (`GL8U16X-GNR2` / `GL8U8X-GNR2 SYN`)

Every chassis class carries the GL prefix. The number and U that follow it are the rack height, so GL4U10G-GNR2 is a 4U chassis.

General Purpose

Single-socket 1U Intel servers for the machines that surround your GPUs: management, services and everyday workloads.

Chassis Class	Height	Drive Bays	Network
`GL1U4LW-C262/2L2T RPSU`	1U	4× 3.5" SATA	2× 1GbE + 2× 10GbE, redundant PSU
`GL1U4LW-C262/2L2T`	1U	4× 3.5" SATA	2× 1GbE + 2× 10GbE
`GL1U4LW-C252`	1U	4× 3.5" SATA	Per configuration
`GL1U2S-C262/FIO`	1U	Per configuration	Per configuration
`GL1U2E-C252`	1U	2× NVMe	Per configuration

GPU Servers (Scale-Out)

PCIe GPU servers with 1 to 10 cards. Choose the GPU, then the host: Intel Xeon 6, 5th Gen Intel Xeon Scalable, or AMD EPYC 9005 / 9004.

GPU	Chassis Class	Host CPU	Best For
RTX PRO 6000 Blackwell Server Edition	`GL4U8G-GNR2`	Intel Xeon 6	Inference, vision, visual computing
H200 NVL	`GL4U10G-EGS2` / `GL4U10G-GNR2` / `GL4U10G-TURIN2`	Xeon Scalable · Xeon 6 · EPYC 9005	Inference and fine-tuning of larger models
H100 NVL	`GL4U10G-EGS2` / `GL4U10G-GENOA2` · `GL4U8G-EGS2` / `GL4U8G-GENOA2` / `GL4U8G-GNR2`	Xeon Scalable · EPYC 9004 · Xeon 6	Inference and fine-tuning
L40S	`GL2U4G-GENOA/M3` · `GL4U10G-EGS2` / `GL4U10G-GENOA2`	EPYC 9004 · Xeon Scalable	Video AI, rendering, 3D
L40	`GL2U4G-GENOA/M3` · `GL4U10G-EGS2` / `GL4U10G-GENOA2`	EPYC 9004 · Xeon Scalable	Rendering, visualization

GPU Servers (Scale-Up)

Eight GPUs on one NVLink domain: NVIDIA HGX platforms for training and very large model inference.

Platform	GPUs	Memory	Chassis Class
HGX B300	8× Blackwell Ultra SXM	Up to 2.1 TB	`GL8U16X-GNR2` / `GL8U16X-GNRAP2` / `GL8U16X-TURIN2`
HGX B200	8× Blackwell SXM	1.4 TB	`GL8U8X-GNR2 SYN` / `GL4U8X-TURIN2/DLC SYN`
HGX H200	8× H200 SXM	1.1 TB	`GL6U8X-EGS2` / `GL6U8X-TURIN2 SYN`

Blackwell HGX platforms carry fifth-generation NVLink: 1.8 TB/s GPU-to-GPU and 14.4 TB/s total per baseboard.

MGX Servers

NVIDIA's modular GPU server design. MGX platforms follow NVIDIA's reference server architecture, on Intel Xeon 6 or AMD EPYC 9005 hosts.

Chassis Class	GPUs Supported	Host CPU
`GL4UXGM-GNR2`	H200 NVL · H100 NVL	Intel Xeon 6
`GL4UXGM-GNR2 CX8`	RTX PRO 6000 Blackwell Server Edition · H200 NVL	Intel Xeon 6
`GL6UXGM-GNR2 CX8/DLC`	Per configuration	Intel Xeon 6
`GL4UXGM-TURIN2`	H200 NVL · L40S	AMD EPYC 9005
`GL4UXGM-TURIN2 CX8`	RTX PRO 6000 Blackwell Server Edition	AMD EPYC 9005
`MECAI-GH200`	NVIDIA GH200 Grace Hopper	NVIDIA Grace

Storage Servers

Storage that keeps the GPUs fed, in three tiers: all-flash for hot data, hybrid, and capacity for cold data.

Tier	Role	Single Socket	Dual Socket
All-flash	Hot tier	`GL1U12E-GENOA` / `GL1U24E1S-GENOA` / `GL1U8E1S-GNRDNO`	`GL1U12E-GNR2/EVAC` / `GL2U24E-GENOA2` / `GL2U16E3S-GNR2FLW`
Hybrid	Flash and disk	`GL1U4L4E-GENOA` / `GL1U8S4E-GENOA` / `GL1U12L4E-GENOA/2Q`	`GL1U4L4E-GNR2` / `GL2U12L4E-2S-GNR2` / `GL2U12L8E-GENOA2`
Capacity	Cold and bulk	`GL1U4L-GENOA/2T` / `GL4U36L6E-GENOA` / `GL4U60L8E-GENOA`	—

Bay counts for three keys: GL1U12E = 12 U.2 NVMe. GL4U36L6E = 36 HDD + 6 NVMe. GL4U60L8E = 60 HDD + 8 NVMe.

Multi-Node Servers

More nodes per rack unit: several independent nodes in one chassis, for dense clusters and CPU-bound workloads.

Chassis Class	Nodes	Density	Platform
`GL2U4N-F/GENOA-M3`	4	2 nodes/U	AMD EPYC 9004, front I/O
`GL1U2N2G-C252/2T`	2, one GPU each	2 nodes/U	Intel C252
`GL1U2N2G-AM5/2T`	2, one GPU each	2 nodes/U	AMD Ryzen (AM5)
`GL4U18N-C266/2T`	18	4.5 nodes/U	Intel C266
`GL4U18N-C252/2T`	18	4.5 nodes/U	Intel C252

GPU and platform names are trademarks of their respective owners.

Tell us what you're building.

A solutions architect responds directly — no ticket queue.

Talk to a Solutions Architect