Home
Compare GPUs
MI355X vs GB200

AMD MI355X vs Nvidia GB200

Two data center GPUs, one CDNA 4 and one Grace Blackwell: 288 GB HBM3e against 186 GB HBM3e.

At a glance

Memory

MI355X: 288 GB
GB200: 186 GB

1.5x more on MI355X

Memory bandwidth

MI355X: 8.00 TB/s
GB200: 8.00 TB/s

Identical

Cheapest on-demand

MI355X: $8.60 /GPU/hr
GB200: $10.50 /GPU/hr

18% cheaper on MI355X

Price comparison

Median price per GPU per hour by billing type, with the cheapest listing under it.

Billing	MI355X	GB200
On-demand	-- /GPU/hr from $8.60	$16.00 /GPU/hr from $10.50
Reserved	No reserved listings	-- /GPU/hr from $10.58
Spot	-- /GPU/hr from $4.50	-- /GPU/hr from $27.04

Heads up: A provider's own page may quote a different figure, for example on tax, region or a promotion. A monthly-only plan shows a derived hourly rate. Medians are across providers, one vote each, so a large catalog does not outweigh a small one. No median for on-demand and spot on the MI355X and reserved and spot on the GB200: we only show one when at least 3 providers list the GPU on that billing type. Verify before provisioning. More on how we price.

Price history

MI355X vs GB200 price history

Weekly median on-demand price per GPU per hour

Last 12 months

Aggregating historical prices...

Specs comparison

Spec	MI355X	GB200	Difference
Memory
Capacity	288 GB HBM3e	186 GB HBM3e	1.5x more on MI355X
Bandwidth	8,000 GB/s	8,000 GB/s	Identical
Bus width	8,192-bit
Compute · dense
FP4	10,066.3 TFLOPS	10,000 TFLOPS	Within 5%
FP8	5,033.2 TFLOPS	5,000 TFLOPS	Within 5%
FP16 / BF16	2,516.6 TFLOPS	2,500 TFLOPS	Within 5%
INT8	5,033.2 TOPS	5,000 TOPS	Within 5%
FP32	157.3 TFLOPS	80 TFLOPS	2x higher on MI355X
FP64	78.6 TFLOPS	40 TFLOPS	2x higher on MI355X
Platform
Interconnect	Infinity Fabric 1,075 GB/s	NVLink 1,800 GB/s	1.7x higher on GB200
Board power	1,400 W	1,200 W	1.2x higher on MI355X
Released	Q2 2025	Q1 2024	15 months newer on MI355X
Precision
Supported formats	FP4FP6FP8FP16BF16FP32FP64INT8	FP4FP6FP8FP16BF16TF32FP32FP64INT8

Compute: Vendor peak figures, dense and per GPU. Not measured throughput.

Source: MI355X datasheet and GB200 datasheet.

Where to rent them

Providers listing these GPUs, those with both first, so you can switch without moving clouds.

Provider	MI355X /GPU/hr	GB200 /GPU/hr
Oracle Cloud	On-Demand $8.60	On-Demand $16.00
DigitalOcean	Spot $4.50	--
TensorWave	Custom $2.95	--
Vultr	Spot $2.59	--
CoreWeave	--	On-Demand $10.50
Azure	--	On-Demand $27.04
AWS	--	Reservation $10.58
CUDO	--	Custom on request
Canopy Wave	--	Custom on request
Nebius	--	Custom on request
Together AI	--	Custom on request