AMD Threadripper Halo Station is probably the world's most expensive desktop PC thanks to its 2.7TB memory and four AI GPUs — a GB300 rival likely to cost $150,000 with just 1 CPU
TechRadar

AMD Threadripper Halo Station is probably the world's most expensive desktop PC thanks to its 2.7TB memory and four AI GPUs - a GB300 rival likely to cost $150,000 with just 1 CPU

AMD Threadripper Halo Station: A Desktop PC Like No Other

AMD's Threadripper Halo Station combines massive memory, four AI accelerators, and 96 CPU cores for demanding local AI workloads.

The Most Expensive Desktop PC?

AMD Threadripper Halo Station is probably the world's most expensive desktop PC thanks to its 2.7TB memory and four AI GPUs - a GB300 rival likely to cost $150,000 with just 1 CPU.

The Halo Station's Key Features

  • The device combines 96 CPU cores with four powerful AI accelerators.
  • Up to 2.6TB of memory gives enormous AI models more local headroom.
  • The Ryzen Threadripper PRO 9995WX processor is based on AMD's Zen 5 architecture and carries 192 threads.

Data Center-Class Computing in a Desktop

Unlike conventional workstations, AMD is building this system around data center-class computing hardware inside a single deskside machine.

Accelerator Configuration

  • The accelerator configuration can use four Instinct MI350P units, with each device carrying 144GB of HBM3e memory and 4TB/s bandwidth.
  • The system's CPU and accelerator memory can reach 2.6TB, giving large AI models far more room to run locally.
  • AMD claims up to 16.4TB/s of combined memory bandwidth, giving large datasets rapid access to both system and accelerator memory.

Comparison with Nvidia's DGX Station

  • The Halo Station compares with Nvidia's DGX Station, whose GB300 platform offers 748GB of coherent memory across its CPU and GPU resources.
  • The comparison shows why AMD is focusing heavily on memory capacity, rather than treating processor core count as the main selling point.

Use Cases for the Halo Station

  • The workstation is intended for researchers, model developers, engineers, creators, and organizations that need intensive workloads handled locally.
  • Its main role is running, training, and fine-tuning large AI models that would normally require access to expensive remote computing infrastructure.
  • The system can handle frontier-class models while supporting hundreds of AI agents operating within the same local environment.

Availability and Pricing

  • AMD unveiled a prototype at IFA 2026, with commercial availability planned for 2027.
  • Industry estimates suggest fully equipped systems could cost around $150,000, although AMD has not announced an official retail price.
Read on TechRadar ↗ ← Back to News

Comments

No comments yet. Start the discussion.