LM TEK RM 4U8G 8 GPU Liquid Cooled Server Chassis

The LM TEK RM 4U8G 8 GPU Liquid Cooled Server Chassis is a high-density, barebone 4U rackmount platform engineered for demanding AI, HPC and GPU compute workloads, supporting up to eight full-height PCIe Gen 5 GPUs within a standard 19 inch rack chassis. Its integrated closed-loop liquid cooling system combines a pump, manifold, GPU and CPU water blocks, radiator and reservoir within the chassis, eliminating the need for an external CDU or facility plumbing while providing efficient cooling for high-power configurations. Designed around standard EATX and EEB server motherboards, the platform avoids proprietary motherboard lock-in and supports off-the-shelf enterprise CPUs, memory, NVMe storage and networking hardware. A platform-level power distribution system with five hot-swappable 2000W 80 PLUS Titanium redundant PSUs provides up to 8000W of sustained chassis power, while each GPU slot supports up to 700W sustained. Measuring 760 x 482 x 178mm and weighing approximately 38kg in barebone form, the LM TEK 4U8GPU10KW provides an exceptionally dense and flexible foundation for AI inference, model development, fine-tuning, research and other multi-GPU workloads.

For business enquiries, contact our B2B sales team here.

Features

---

LM TEK 4U8GPU10KW AI & HPC Server Platforms

A Barebone Server Platform Engineered as a Single Thermal and Power System

A barebone AI server platform: eight full-height PCIe Gen 5 GPU slots in a 4U chassis, engineered around a closed liquid cooling loop with an in-chassis radiator, so it deploys in a standard server room without facility plumbing or an external CDU. Platform-level power, no motherboard lock-in. Cooling designed by the EK engineering team; engineered and assembled in Komenda, Slovenia.

A barebone platform: LM TEK supplies the chassis, the integrated cooling loop, and platform-level power. You specify the board, CPUs, GPUs, memory, and storage.

LM TEK 4U8GPU10KW is the result of asking what an 8-GPU AI server should look like if you started over. The chassis is engineered around a closed liquid cooling loop with an in-chassis radiator. Power distribution is platform-level, separated from the motherboard's VRM. The PCIe slot layout is built around the cooling loop's physical envelope, not retrofitted around a board form factor.

What the platform does not do is force you onto a proprietary motherboard. The board, the CPUs, the memory, the storage, the network adaptors: all off-the-shelf enterprise parts, sourced and refreshed independently of the chassis. The reason customers buy LM TEK 4U8GPU10KW is the cooling and the chassis discipline; the reason customers stay with LM TEK 4U8GPU10KW is that nothing else about the build is locked to one supplier.

Engineered and assembled in Komenda, Slovenia. Cooling system designed by the EK engineering team, twenty-five years deep in high-performance liquid cooling. Pressure envelope, materials, and operating range are documented at the supplier level and do not differ from the rest of the LM TEK liquid-cooling catalogue.

Six Things that Matter

4U Envelope

Eight full-height GPU slots in a four-rack-unit chassis. Density that normally needs 6U or 8U.

Eight GPU Slots

Full-height PCIe slots, each provisioned for liquid-cooled GPU water blocks at the platform's power envelope.

Integrated Liquid Cooling

Closed loop, in-chassis radiator, no external CDU. Deploys in standard server rooms without facility plumbing.

No Motherboard Lock-In

Standard EATX and EEB server boards from your vendor of choice. CPU, RAM, NVMe, NICs are all off-the-shelf.

EK-Engineered Cooling

Materials discipline, pressure envelope, and manifold geometry from the EK liquid-cooling reference, transferred to enterprise rack scale.

Platform Model

Barebone chassis with integrated cooling, power, and platform support. GPUs and internals specified per deployment.

Key Engineering - Three Design Decisions Worth Calling Out

E-ATX-Compatible - 4U - 19" Rack

The chassis is engineered around standard EATX server motherboards rather than a proprietary board. PCIe slot positions and clearance accommodate the cooling loop without needing custom riser cards. Mounting hardware is rack-standard. The platform takes ordinary enterprise components and arranges them at extraordinary density.

Closed Integrated Liquid Loop - In-Chassis Radiator

Pump, manifold, GPU and CPU water blocks, radiator, and reservoir are all inside the chassis. The radiator faces the front intake; airflow paths are engineered alongside the loop, not as an afterthought. Working pressure 2500 Pa; maximum pressure 30 000 Pa with EK-Pro PRM, 75 000 Pa otherwise. Materials stack: nickel-plated copper, stainless steel, POM Acetal, EPDM seals.

Redundant High-Density PDU - Platform-Level Distribution

Power distribution is engineered at the platform level so the GPU power budget does not depend on the motherboard's VRM. Redundant PSUs, hot-swap modules, and headroom for the full 8-GPU configuration at maximum sustained load. The customer's motherboard chooses CPU and memory; the platform chooses how power reaches the GPUs.

Proof and Deployment

The Honest Envelope

• Deploys in standard server-room conditions: 5 to 35°C ambient, 5 to 60°C coolant.
• No facility plumbing. No external CDU.
• Five redundant 2000 W 80 PLUS Titanium PSUs, up to 8000 W sustained.
• You still provision rack power and rack cooling headroom.

Plan for the Heat

A dense GPU server converts nearly all of its electrical draw into heat in the room. In extended thermal torture testing, a fully populated LM TEK 4U8GPU10KW measured 6800 to 7000 W sustained draw. That heat load is defined by the GPUs and CPUs in the configuration, the same as any other max-spec 8-GPU server. Without added room cooling, ambient will rise under sustained load. The LM TEK 4U8GPU10KW’s operating envelope extends to 35°C ambient, so a room that warms under load stays inside specification, but scope rack cooling for the configured load before deployment.

In LM TEK torture testing at 30°C ambient, GPU package temperatures held at 76 to 81°C, inside the envelope a real server room can offer. Acoustic measurements under sustained load are available on request.

Efficiency, and the Water Question

Power conversion uses 80 PLUS Titanium supplies, the highest rating in the certification; at multi-kilowatt sustained draw, conversion losses are a real operating cost and an increasingly common bid and tender criterion. And the water objection to AI infrastructure does not apply here: the loop is closed, filled once at assembly with engineered coolant, and consumes no water in operation. There is no facility water connection to consume from.

Reference Configurations

From Starter Footprint to Production Maximum

Reference configurations are examples drawn from real customer builds, not a fixed bill of materials. The final BOM is specified per deployment; LM TEK engineering validates compatibility and cooling capacity for the chosen GPU before quote.

Tier	GPU	CPU	RAM	Storage	Use Case
Modest start	2x NVIDIA RTX 6000 Ada	1x AMD EPYC 9354 (32-core)	256 GB DDR5 ECC	2x 3.84 TB NVMe	Workstation crossover · departmental AI · single-team inference
Mid-tier	4x NVIDIA RTX A6000	2x AMD EPYC 9474F (48-core)	512 GB DDR5 ECC	4x 3.84 TB NVMe	Lab / research group · multi-user inference · light fine-tuning
Production maximum	8x NVIDIA H200 NVL	2x Intel Xeon Platinum 8580	1 TB DDR5 ECC	8x 7.68 TB NVMe	Enterprise inference · production fine-tuning · multi-tenant

Compatibility

Motherboards

EATX, EEB, and server reference boards from major vendors. Workstation boards in the same envelope are also supported but not the primary use case.

CPU Sockets

AMD SP5 (EPYC 9004 / 9005), Intel LGA 4677 (Xeon Scalable 4th / 5th gen) and Intel LGA 4710/Intel LGA 4710-2 verified. Other server sockets supported via available motherboard choices.

GPUs

Full-height PCIe Gen 5, up to 700 W per slot sustained, water-block-equipped. Verified with the GPU water-block catalogue (RTX 6000 Ada, RTX A6000, H100 NVL, H200 NVL, A100, RTX PRO 6000 Blackwell, others on request).

Networking

OCP 3.0 mezzanine and standard PCIe NICs supported.

Compatible Components

Everything in the Loop, from the Same Catalogue

The cooling loop inside LM TEK 4U8GPU10KW is built from the same components that ship into LM TEK workstation deployments. Same materials discipline, same pressure envelope, same supplier.

• EK-Pro GPU WB H200 NVL
• EK-Pro GPU WB RTX 6000 Ada Rack
• EK-Pro GPU WB A100 Rack
• EK-Pro GPU WB H100 NVL Rack
• EK-Pro GPU WB RTX A5500 Rack
• EK-Pro GPU WB Zotac RTX 5090
• EK-Pro GPU WB RTX PRO 6000 Blackwell
• CPU WB STR
• EK-Pro CPU WB SP5 Ni + Acetal
• EK-Pro CPU WB 4677 Ni + Acetal
• EK-Pro CPU WB 4189

For Procurement

The buy reason is the cooling; the stay reason is supplier independence.

• No proprietary components; every internal part has a second source.
• Board and silicon are sourced and refreshed independently of the chassis.
• One platform carries a build from two GPUs to eight 700 W GPUs, so the entry configuration and the end state are the same asset.
• Titanium-rated power conversion and zero operational water use are scoreable criteria in energy-weighted tenders.
• A full Procurement and Total Cost of Ownership brief is available on request.