AMD MI355X vs Nvidia A800

The MI355X is newer, with 7.2x the memory and 5.1x the bandwidth. The A800 is cheaper right now.

At a glance

Memory

7.2x more on MI355X

MI355X: 288 GB

A800: 40 GB

Memory bandwidth

5.1x higher on MI355X

MI355X: 8.00 TB/s

A800: 1.55 TB/s

Cheapest on-demand

80% cheaper on A800

MI355X: $5.99 /GPU/hr

A800: $1.18 /GPU/hr

Price comparison

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

Billing	MI355X	A800
On-demand	$8.04 /GPU/hr from $5.99	-- /GPU/hr from $1.18
Reserved	-- /GPU/hr from $4.29	No reserved listings
Spot	-- /GPU/hr from $3.44	No spot listings

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 reserved and spot on the MI355X and on-demand on the A800: 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 A800 price history

Weekly median on-demand price per GPU per hour

Last 12 months

Aggregating historical prices...

Specs comparison

Spec	MI355X	A800	Difference
Memory
Capacity	288 GB HBM3e	40 GB HBM2	7.2x more on MI355X
Bandwidth	8,000 GB/s	1,555 GB/s	5.1x higher on MI355X
Bus width	8,192-bit	5,120-bit	60% wider on MI355X
Compute · dense
FP4	10,066.3 TFLOPS	Not supported	MI355X only
FP8	5,033.2 TFLOPS	Not supported	MI355X only
FP16 / BF16	2,516.6 TFLOPS	312 TFLOPS	8.1x higher on MI355X
INT8	5,033.2 TOPS	623.5 TOPS	8.1x higher on MI355X
FP32	157.3 TFLOPS	19.5 TFLOPS	8.1x higher on MI355X
FP64	78.6 TFLOPS	9.7 TFLOPS	8.1x higher on MI355X
Platform
Interconnect	Infinity Fabric 1,075 GB/s	NVLink 400 GB/s	2.7x higher on MI355X
Board power	1,400 W	240 W	5.8x higher on MI355X
Released	Q2 2025	Q4 2022	3 years newer on MI355X
Precision
Supported formats	FP4FP6FP8FP16BF16FP32FP64INT8	FP16BF16TF32FP32FP64INT4INT8

Figures are for each model's headline configuration. Where a form factor carries a different number, it is named under the figure.

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

Source: MI355X datasheet and A800 datasheet.

Where to rent them

Providers that rent these GPUs, with each one's cheapest rate. No provider we track offers both right now.

Provider	MI355X	A800
Daytona USA	$5.99 /GPU/hr On-demand	--
OpenRelay USA	$6.30 /GPU/hr On-demand	--
Oracle Cloud USA	$8.60 /GPU/hr On-demand	--
Cirrascale USA	$4.29 /GPU/hr Reserved	--
Impossible Cloud Germany	$6.13 /GPU/hr Reserved	--
DigitalOcean USA	$4.50 /GPU/hr Spot	--
TensorWave USA	$2.95 /GPU/hr Custom OAM	--
Crusoe USA	On request Custom	--
GPUhub Singapore	--	$1.18 /GPU/hr On-demand NVL