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
SpecAWS Trainium2El Capitan
Accelerator—AMD Instinct MI300A (APU)
Bf16 Tflops PER Chip667—
Chips PER Trn2 Instance16—
Chips PER Ultraserver64—
Cores—11,340,000
CPU—AMD 4th Gen EPYC 24C 1.8GHz
FP8 Dense Petaflops PER Chip1.3—
FP8 Sparse Petaflops PER Chip5.2—
HBM Bandwidth TB S PER Chip2.9—
HBM Capacity GB PER Chip96—
HBM TypeHBM3—
Hpcg Tflops—17,406
Installation Year—2,024
Interconnect—HPE Slingshot-11
Location—DOE/NNSA/Lawrence Livermore National Laboratory
Manufacturer—HPE
Model—HPE Cray EX255a
Neuroncores PER Chip8—
Power KW—29,684.62
Rmax Pflops—1,809
Rpeak Pflops—2,821.1
Top500 Jun2026 NO1—LineShine (National Supercomputing Centre in Shenzhen, China)
Top500 Rank Jun2026—2
Top500 Rank Nov2024—1
Top500 Rank Nov2025—1
Trainium3 GA2025-12-02—
Trainium3 HBM Bandwidth TB S4.9—
Trainium3 Hbm3e GB PER Chip144—
Trn2 Instance FP8 Petaflops20.8—
Trn2 Instance HBM TB1.5—
Trn3 Ultraserver FP8 Pflops362—
Trn3 Ultraserver MAX Chips144—
Ultraserver FP8 Petaflops83.2—
Ultraserver Total HBM Bandwidth TB S185—
Ultraserver Total HBM TB6—