Rack-scale demonstration Deployment

SoftBank–Foxconn H200 rack cooling pilot

SoftBank, ZutaCore, and Foxconn demonstrated a rack-integrated two-phase direct-to-chip system using NVIDIA H200 GPUs.

NVIDIA H200SoftBankFoxconntwo phase

Project brief

What is publicly known

The companies reported an operational demonstration and performance evaluation at a SoftBank data center in February 2025. The rack used ZutaCore HyperCool and was reported to pass NVIDIA temperature testing with a measured partial PUE of 1.03 for the rack cooling setup.

Status is labeled from public evidence. Supplier orders, MoUs, pilots, projects under delivery, and operating sites are not counted as equivalent.

Status
Pilot
Operator
SoftBank
Location
Japan
Capacity
One rack-scale demonstration
Rack density
Not disclosed
Compute
NVIDIA H200 GPUs
Announced
May 21, 2025

Connected records

Companies involved

Connected records

Cooling architecture

Connected records

Equipment chain

Evidence ledger

Primary sources

Profiles distinguish company-reported claims from confirmed operating data. Open the original materials before relying on a specification.

  1. 01

Direct answers

Frequently asked questions

Concise answers generated from the structured record and its cited evidence.

What is the status of SoftBank–Foxconn H200 rack cooling pilot?

SoftBank–Foxconn H200 rack cooling pilot is tracked as pilot. It was announced on May 21, 2025, and the public record should not be interpreted as a later project stage without new evidence.

How much capacity does SoftBank–Foxconn H200 rack cooling pilot represent?

The source-reported capacity is One rack-scale demonstration. The project is located in Japan.

Which cooling technologies are associated with SoftBank–Foxconn H200 rack cooling pilot?

The record connects the project to Two Phase Direct To Chip.