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

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

Rig A

NVIDIA · workstation

Change

Rig B

AWS · cloud-model

Change

0 of 40 fields comparable

Spec	NVIDIA GB200 NVL72	AWS Trainium2
Bf16 Tflops PER Chip	—	667
Blackwell Gpus	72	—
Chips PER Trn2 Instance	—	16
Chips PER Ultraserver	—	64
Compute Trays	18	—
Cooling	liquid	—
CPU Memory Bandwidth TB S	14	—
CPU Memory Lpddr5x TB	17	—
Fp16 Bf16 Pflops	360	—
FP32 TFLOPS	5,760	—
FP64 TFLOPS	2,880	—
FP8 Dense Petaflops PER Chip	—	1.3
FP8 FP6 Pflops	720	—
FP8 Sparse Petaflops PER Chip	—	5.2
GPU Memory Bandwidth TB S	576	—
Grace Cpus	36	—
HBM Bandwidth TB S PER Chip	—	2.9
HBM Capacity GB PER Chip	—	96
HBM Type	—	HBM3
Int8 Pops	720	—
Neuroncores PER Chip	—	8
Nvfp4 Pflops Dense	720	—
Nvfp4 Pflops Sparse	1,440	—
Nvlink Generation	5	—
Nvswitch Trays	9	—
Rack Power KW Approx	120	—
Tf32 Pflops	180	—
Total ARM Cores	2,592	—
Total Hbm3e TB	13.4	—
Total Nvlink Bandwidth TB S	130	—
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