Uvation MarketplaceMarketplace

Supermicro SYS-222GS-NB3OT-ALC GPU Server

Liquid-cooled 2 OU server pairing eight NVIDIA Blackwell Ultra GPUs with dual Intel Xeon 6700-series CPUs for 21-inch OCP ORv3 racks.

GPUs
8x
GPUs
HBM3e
2.3 TB
HBM3e
OCP ORv3
2 OU
OCP ORv3

Eight-GPU Compute Sized for Open Racks

Supermicro’s SYS-222GS-NB3OT-ALC packs eight Blackwell Ultra GPUs and dual Intel Xeon processors into a liquid-cooled 2 OU node built for OCP ORv3 racks.

Full HGX B300 Power in 94 mm

Direct-to-chip cold plates cool the CPUs and GPUs, fitting eight Blackwell Ultra GPUs into a 94 mm-tall chassis.

32 DIMMs to Feed Data-Hungry Models

Up to 8 TB of DDR5 across 32 DIMMs and 8 hot-swap E1.S NVMe bays keep datasets close.

Faster Attention for Reasoning-Heavy AI

Blackwell Ultra GPUs deliver 2x the attention performance of HGX B200, accelerating reasoning models, long-context inference, and large-scale training.

Eight GPUs. One Fabric. Two OU.

Fifth-generation NVLink and NVSwitch connect all eight GPUs into one unified memory and compute domain.

Supermicro SYS-222GS-NB3OT-ALC 2 OU chassis
8× GPUs
NVIDIA HGX B300 Blackwell Ultra
2.3 TB
HBM3e GPU memory
1.8 TB/s
NVLink bandwidth per GPU
2 OU / D2C
Direct-to-chip liquid cooled

A platform for models that refuse to fit the old shape.

From frontier training to real-time reasoning, this liquid-cooled server runs demanding AI and HPC workloads.

Train Frontier Models in Fewer Days

72 petaFLOPS of sparse FP8 compute and 2.3 TB of HBM3e help teams train larger models in less time.

Low-Latency Inference for Agentic AI

108 petaFLOPS of dense FP4 compute speed up agentic reasoning and long-context responses.

Simulation, Genomics and Research Computing

Dual Xeon CPUs with up to 172 cores and 8 TB of DDR5 feed simulation, genomics, research and analytics alongside AI.

Engineered to dominate

Cold-Plate Cooled, Rack-Scale Ready

Direct-to-chip cooling sustains Blackwell Ultra compute, NVLink 5 bandwidth and 800 Gb/s networking per GPU in a dense 2 OU node.

Exploded view of the Supermicro SYS-222GS-NB3OT-ALC
144 PFLOPS
FP4 sparse (108 dense)
72 PFLOPS
FP8/FP6 sparse compute
2×
Attention vs. HGX B200
800 Gb/s
8× ConnectX-8 ports
Exploded view of the Supermicro SYS-222GS-NB3OT-ALC
144 PFLOPS
FP4 sparse (108 dense)
72 PFLOPS
FP8/FP6 sparse compute
2×
Attention vs. HGX B200
800 Gb/s
8× ConnectX-8 ports

The rack, itemized.

SUBSYSTEM

SPECIFICATION / CONFIGURATION DETAILS

Processor

2× Intel Xeon 6700P; up to 86 cores each

Operating System

RHEL 10.0

Chipset

System on Chip (SoC)

Accelerators / GPU

NVIDIA SXM: HGX B300 8-GPU (288GB)

Memory

32 DIMMs; up to 4 TB or 8 TB, depending on speed

Primary Storage

8 × front hot-swap E1.S NVMe drive bays

Secondary Storage

2 × M.2 NVMe boot drives

Security

Trusted Platform Module (TPM) 2.0; Silicon Root of Trust (RoT) – NIST 800-193 Compliant

Chassis

CSE-GP202TS-000NP

Power Supply

Power via Busbar (48-52V DC)

Ports

8× OSFP; 2× 10GbE; 1GbE BMC; USB; VGA

Slots / PCIe Architecture

2 PCIe 5.0 x16 (in x16) FHHL slots

Bezel

N/A

Form Factor

2 OU Rackmount

Network Card

8x NVIDIA ConnectX®-8 SuperNICs (up to 800Gb/s)

Warranty

3-Year Labor, 1-Year Parts & 1-Year Advance Cross-Shipment Limited Manufacturer Warranty

Front view of the Supermicro SYS-222GS-NB3OT-ALC GPU server

Busbar Power Meets 800 Gb/s Networking

48–52 V DC busbar power delivery

Shared rack power shelves feed the node through a 1400A ORv3 busbar instead of in-server power supplies.

8× 800 Gb/s ConnectX-8 SuperNIC ports

Eight 800 Gb/s ConnectX-8 SuperNICs give every GPU its own path to fast, low-latency multi-node scale-out.

2× PCIe 5.0 x16 FHHL expansion slots

Two full-x16 PCIe 5.0 slots support up to two more NICs for north-south network traffic.

Nuclear-Ready, Plug and Play AI Factories

Bypass grid constraints with self-contained, high-density inference ready infrastructure directly adjacent to clean, base load nuclear power.

Global Deployment Engine

Any Infra, Anywhere in the world.

Global Shipping and Deployment Capability treats advanced infrastructure as mobile, modular freight.

Packaged complex computing nodes, cooling manifolds, and power distribution units into certified, standard form factors.

360° Service Ecosystem

Unifying Legacy with Next-Gen Compute.

Vertically integrated service ecosystem that serves as an operational bridge between legacy environments and next-generation, AI-native infrastructure.

Finance and Leasing

Capital-as-a-Service for High-Density Compute

Convert upfront capital expenditures (CapEx) into predictable operating expenses (OpEx).

Uvation allows enterprises and sovereign entities to deploy multi-megawatt AI factories and advanced GPU clusters without liquidity constraint.