Skip to content
MCS · Rigs · Compare

Side-by-side, with the numbers that matter.

Two rigs, one table. Higher numbers win in cobalt; TDP and price flip.

Rig A

The Stargate Project

OpenAI / Oracle / SoftBank / MGX · workstation

Change

Rig B

NVIDIA DGX H200

NVIDIA · workstation

Change

0 of 33 fields comparable

⇄ swap sides
SpecThe Stargate ProjectNVIDIA DGX H200
Abilene Trained ModelGPT-5.5—
Abilene Trained Model Version5.5—
Capacity Added Prior 90 Days GW MIN3—
Capacity Added Window Days90—
Capacity Update Date2026-04-29—
CPU—2x Intel Xeon Platinum 8480C
CPU Cores Total—112
Custom Silicon AnnouncedBroadcom (10 GW)—
FP8 Petaflops—32
GPU Count—8
GPU Model—H200 SXM
Internal Storage—8x 3.84TB NVMe U.2
MAX Power KW—10.2
Network MAX Gbps PER Port—400
Network Peak TB S—1
Network Ports—8x ConnectX-7 single-port (4x OSFP)
Nvlink PER GPU Bandwidth GB S—900
Nvswitch Bandwidth TB S—7.2
Nvswitch Count—4
Oracle Partnership GW5—
OS Storage—2x 1.92TB NVMe M.2
PartnersOpenAI, SoftBank, Oracle, MGX—
Planned Capacity GW 20257—
Planned Chip Count Million2—
Primary Accelerator AnnouncedNVIDIA GB200 (initial Oracle racks)—
Primary SiteAbilene, Texas—
Secondary Accelerator AnnouncedAMD Instinct (up to 6 GW)—
Secured Capacity Commitment Deadline Year2,029—
Secured Capacity Commitment GW10—
Status Checked2026-09-24—
System Memory TB—2
Total GPU Memory GB—1,128
Total Planned Investment USD Billion400—