LiquidMax RackScale 64
LiquidMax® RackScale 64

High-Density Liquid Cooled AI

Fully liquid-cooled 42U rack architecture designed to support 64 GPUs, engineered for compact, high-performance AI infrastructure.

LiquidMax RackScale 128
Left Aligned Gradient Text with Divider
Key Features
AMAX Specialized IT Solutions
Management Control System

64x GPU Capacity

Supports up to 64x NVIDIA Blackwell GPUs interconnected with NVIDIA NVLink™ across 8x 4U liquid-cooled servers.

Enhanced Cooling Capacity

Liquid Cooling

Direct-to-chip liquid cooling with CDU, hot-swap pump support, and vertical cooling manifold built into the rack.

Streamlined Servicing

All-in-One Rack Solution

Pre-assembled with compute, power, and cooling. Shipped fully tested and ready for immediate deployment.

Reduced Power Usage

Compact AI Infrastructure

Delivers up to 11.5TB of HBM3e memory in a single 42U footprint for AI training, inference, and large-scale model development.

Left Aligned Gradient Text with Divider
System Architecture
RackScale Overlay Diagram
400G Networking
8x 4U Servers
Vertical PDUs
Vertical Rack Manifold
Coolant Distribution
Unit (CDU)

Thermal Design and Reliability

The LiquidMax® RackScale 64 is built for efficient, high-density AI compute. Each 42U rack integrates 8x 4U liquid-cooled servers using the NVIDIA HGX™ B200 platform, delivering up to 64 GPUs interconnected via NVIDIA NVLink. This architecture supports large-scale model training and inference with 11.5TB of HBM3e memory per rack.

Engineered with direct liquid cooling, RackScale 64 improves thermal efficiency while lowering Power Usage Effectiveness (PUE). The system eliminates traditional server fans by using passive cold plates and a high-precision CDU with dual hot-swappable pumps. This design reduces energy loss, maintains stable temperatures under heavy GPU loads, and contributes to longer hardware lifespan, all within a compact footprint.

Left Aligned Gradient Text with Divider
Why AMAX

Engineering Expertise

Our team of thermal, electrical, mechanical, and networking engineers are skilled in designing solutions designed to your specific requirements.

AMAX Specialized IT Solutions
Solution Architect

Solution Architects

AMAX's solution architects optimize IT configurations for performance, scalability, and industry-specific reliability.

Networking Solutions

Networking

AMAX designs custom networking topologies to enhance connectivity and performance in AI and HPC environments.

Thermal Management

Thermal Management

AMAX implements innovative cooling technologies that boost performance and efficiency in dense computing setups.

Benchmarking and Optimization

Compute Optimization

AMAX ensures maximum performance through benchmarking and testing, aligning hardware and software for AI workloads.

Left Aligned Gradient Text with Divider
Server Design

Liquid-Cooled Performance at Scale

The AceleMax AXG-428IL pairs 8x NVIDIA Blackwell GPUs with dual Intel Xeon 6700-series CPUs, PCIe Gen 5 NVMe storage, and 10 expansion slots. Designed for dense AI deployments, it delivers high-speed interconnects and efficient thermal control in a 4U rackmount form factor.

AceleMax AXG-428IL Liquid Cooled B200 GPU Server

Left Aligned Gradient Text with Divider
AI Architects
Design to Deployment

From Design to Deployment

AMAX's approach to AI solutions begins with intelligent design, emphasizing the creation of high-performance computing and network infrastructures tailored to AI applications. We guide each project from concept to deployment, ensuring systems are optimized for both efficiency and future scalability.

Left Aligned Gradient Text with Divider
HostMax
Networking Text and Image

Activate Your Infrastructure Instantly with HostMax™

HostMax™ is AMAX’s in-house deployment service that allows you to power on and operate your liquid-cooled AI systems the moment they’re built. Instead of waiting for external colocation space, HostMax™ provides immediate hosting within AMAX’s facility. This enables a direct transition from assembly to deployment for testing, validation, and early production use.

Learn More
hostmax hosting area render
Left Aligned Gradient Text with Divider
LiquidMax Rackscale 128

LiquidMax® RackScale 64

The LiquidMax® RackScale 64 is a high-density, fully liquid-cooled 42U rack solution designed to support advanced AI training and inference.

  • Supports up to 64x NVIDIA Blackwell GPUs
  • 8x 4U liquid-cooled servers with NVIDIA NVLink interconnect
  • Integrated CDU and vertical cooling distribution manifold
  • Rack-scale solution with pre-integrated compute, cooling, and power
RackScale 128 Feature
Left Aligned Gradient Text with Divider
Request a Quote
Form with Responsive Styled Box