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 42 fields comparable

⇄ swap sides
SpecNVIDIA RTX PRO 5000 BlackwellAWS Trainium2
AI Tops2,064—
Bandwidth GB S1,344—
Bf16 Tflops PER Chip—667
Chips PER Trn2 Instance—16
Chips PER Ultraserver—64
Cuda Cores14,080—
Displayport Count4—
Displayport Version2.1—
ECCtrue—
Form factor4.4 in H x 10.5 in L, dual slot—
FP32 TFLOPS65—
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
Neuroncores PER Chip—8
Nvdec Count3—
Nvdec GEN6—
Nvenc Count3—
Nvenc GEN9—
Pcie GEN5—
RT Cores110—
RT Cores GEN4—
RT Tflops196—
Slot Width2—
TDP (W)↓ better300—
Tensor Cores440—
Tensor Cores GEN5—
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)48—
Vram GB ALT72—
Vram TypeGDDR7 ECC—