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

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

Rig A

NVIDIA · gpu

Change

Rig B

AWS · cloud-model

Change

0 of 45 fields comparable

Spec	NVIDIA GeForce RTX 5080 Founders Edition	AWS Trainium2
AI Tops	1,801	—
Bandwidth GB S	960	—
Base Clock GHZ	2.3	—
Bf16 Tflops PER Chip	—	667
Boost Clock GHZ	2.62	—
Chips PER Trn2 Instance	—	16
Chips PER Ultraserver	—	64
Cuda Cores	10,752	—
Displayport Count	3	—
Displayport Version	2.1b	—
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
Hdmi Count	1	—
Hdmi Version	2.1b	—
Length MM	304	—
Memory BUS Bits	256	—
Memory Speed Gbps	30	—
MSRP (USD)↓ better	999	—
Neuroncores PER Chip	—	8
Pcie GEN	5	—
RT Core Tflops	171	—
RT Cores GEN	4	—
Slot Width	2	—
Status Checked	2026-09-24	—
Street Price Date	2026-09-01	—
Street Price Premium PCT US	55	—
Street Price USD US Typical	1,550	—
TDP (W)↓ better	360	—
Tensor Cores GEN	5	—
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
VRAM (GB)	16	—
Vram Type	GDDR7	—
Width MM	137	—