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

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

Rig A

NVIDIA · accelerator

Change

Rig B

AWS · cloud-model

Change

0 of 28 fields comparable

Spec	NVIDIA B100	AWS Trainium2
Bf16 Tflops PER Chip	—	667
Chip TO Chip Interconnect TB S	10	—
Chips PER Trn2 Instance	—	16
Chips PER Ultraserver	—	64
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 GB Unverified	192	—
HBM Type	—	HBM3
Memory Bandwidth TB S Unverified	8	—
Neuroncores PER Chip	—	8
Nvlink Generation	5	—
Nvlink GPU TO GPU TB S	1.8	—
Process Node	TSMC 4NP	—
Status	not listed by NVIDIA	—
TDP Watts Unverified	700	—
Trainium3 GA	—	2025-12-02
Trainium3 HBM Bandwidth TB S	—	4.9
Trainium3 Hbm3e GB PER Chip	—	144
Transistor Count Billion	208	—
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