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.

0 of 35 fields comparable

⇄ swap sides
SpecEl CapitanAWS Trainium2
AcceleratorAMD Instinct MI300A (APU)—
Bf16 Tflops PER Chip—667
Chips PER Trn2 Instance—16
Chips PER Ultraserver—64
Cores11,340,000—
CPUAMD 4th Gen EPYC 24C 1.8GHz—
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
Hpcg Tflops17,406—
Installation Year2,024—
InterconnectHPE Slingshot-11—
LocationDOE/NNSA/Lawrence Livermore National Laboratory—
ManufacturerHPE—
ModelHPE Cray EX255a—
Neuroncores PER Chip—8
Power KW29,684.62—
Rmax Pflops1,809—
Rpeak Pflops2,821.1—
Top500 Jun2026 NO1LineShine (National Supercomputing Centre in Shenzhen, China)—
Top500 Rank Jun20262—
Top500 Rank Nov20241—
Top500 Rank Nov20251—
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