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

AWS Trainium2

AWS · cloud-model

Change

0 of 36 fields comparable

⇄ swap sides
SpecThe Stargate ProjectAWS Trainium2
Abilene Trained ModelGPT-5.5—
Abilene Trained Model Version5.5—
Bf16 Tflops PER Chip—667
Capacity Added Prior 90 Days GW MIN3—
Capacity Added Window Days90—
Capacity Update Date2026-04-29—
Chips PER Trn2 Instance—16
Chips PER Ultraserver—64
Custom Silicon AnnouncedBroadcom (10 GW)—
FP8 Dense Petaflops PER Chip—1.3
FP8 Sparse Petaflops PER Chip—5.2
HBM Bandwidth TB S PER Chip—2.9
HBM Capacity GB PER Chip—96
HBM Type—HBM3
Neuroncores PER Chip—8
Oracle Partnership GW5—
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—
Total Planned Investment USD Billion400—
Trainium3 GA—2025-12-02
Trainium3 HBM Bandwidth TB S—4.9
Trainium3 Hbm3e GB PER Chip—144
Trn2 Instance FP8 Petaflops—20.8
Trn2 Instance HBM TB—1.5
Trn3 Ultraserver FP8 Pflops—362
Trn3 Ultraserver MAX Chips—144
Ultraserver FP8 Petaflops—83.2
Ultraserver Total HBM Bandwidth TB S—185
Ultraserver Total HBM TB—6