AMD MI455X vs Nvidia GH200

The MI455X is newer, with 4.5x the memory and 5.8x the bandwidth.

At a glance

Memory

4.5x more on MI455X

MI455X: 432 GB

GH200: 96 GB

Memory bandwidth

5.8x higher on MI455X

MI455X: 23.30 TB/s

GH200: 4.00 TB/s

Cheapest on-demand

MI455X is not sold on demand

MI455X: --

GH200: $3.87 /GPU/hr

Price comparison

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

Billing	MI455X	GH200	Difference
On-demand	No on-demand listings	-- /GPU/hr from $3.87
Spot	No spot listings	-- /GPU/hr from $1.99

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 GH200: 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

MI455X vs GH200 price history

Weekly median on-demand price per GPU per hour

Last 12 months

Aggregating historical prices...

Specs comparison

Spec	MI455X	GH200	Difference
Memory
Capacity	432 GB HBM4	96 GB HBM3	4.5x more on MI455X
Bandwidth	23,300 GB/s	4,000 GB/s	5.8x higher on MI455X
Compute · dense
FP4	40,265 TFLOPS	Not supported	MI455X only
FP8	20,133 TFLOPS	1,979 TFLOPS	10x higher on MI455X
FP16 / BF16	5,033 TFLOPS	989.5 TFLOPS	5.1x higher on MI455X
INT8	5,033 TOPS	1,979 TOPS	2.5x higher on MI455X
FP32	315 TFLOPS	67 TFLOPS	4.7x higher on MI455X
FP64	Not supported	34 TFLOPS	GH200 only
Platform
Interconnect	Infinity Fabric 3,600 GB/s	NVLink 900 GB/s	4x higher on MI455X
Released	Q3 2026	Q2 2023	3 years newer on MI455X
Precision
Supported formats	FP4FP6FP8FP16BF16FP32INT8	FP8FP16BF16TF32FP32FP64INT8

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

Source: MI455X datasheet and GH200 datasheet.

Where to rent them

Providers that rent these GPUs, with each one's cheapest rate. Providers that rent both are listed first.

Provider	MI455X	GH200
Vultr USA	On request Custom	$1.99 /GPU/hr Spot
Azure USA	On request Custom	--
Cirrascale USA	On request Custom	--
Oracle Cloud USA	On request Custom	--
TensorWave USA	On request Custom	--
Denvr Canada	--	$3.87 /GPU/hr On-demand
CoreWeave USA	--	$6.50 /GPU/hr On-demand
Lambda Labs USA	--	$2.29 /GPU/hr On-demand
Sesterce France	--	$2.52 /GPU/hr On-demand