AMD MI455X vs Nvidia A4000

The MI455X is newer, with 27x the memory and 52x the bandwidth.

At a glance

Memory

27x more on MI455X

MI455X: 432 GB

A4000: 16 GB

Memory bandwidth

52x higher on MI455X

MI455X: 23,300 GB/s

A4000: 448 GB/s

Cheapest on-demand

MI455X is not sold on demand

MI455X: --

A4000: $0.21 /GPU/hr

Price comparison

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

Billing	MI455X	A4000	Difference
On-demand	No on-demand listings	$0.82 /GPU/hr from $0.21
Reserved	No reserved listings	$0.25 /GPU/hr from $0.11

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. Verify before provisioning. More on how we price.

Price history

MI455X vs A4000 price history

Weekly median on-demand price per GPU per hour

Last 12 months

Aggregating historical prices...

Specs comparison

Spec	MI455X	A4000	Difference
Memory
Capacity	432 GB HBM4	16 GB GDDR6	27x more on MI455X
Bandwidth	23,300 GB/s	448 GB/s	52x higher on MI455X
Bus width		256-bit
Compute · dense
FP4	40,265 TFLOPS	Not supported	MI455X only
FP8	20,133 TFLOPS	Not supported	MI455X only
FP16 / BF16	5,033 TFLOPS	76.7 TFLOPS	66x higher on MI455X
INT8	5,033 TOPS
FP32	315 TFLOPS	19.2 TFLOPS	16x higher on MI455X
Platform
Interconnect	Infinity Fabric 3,600 GB/s	PCIe only	Infinity Fabric on MI455X only
Released	Q3 2026	Q2 2021	5 years newer on MI455X
Precision
Supported formats	FP4FP6FP8FP16BF16FP32INT8	FP16BF16TF32FP32INT4INT8

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

Source: MI455X datasheet and A4000 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	A4000
Cirrascale USA	On request Custom	$0.28 /GPU/hr Reserved
Azure USA	On request Custom	--
Oracle Cloud USA	On request Custom	--
TensorWave USA	On request Custom	--
Vultr USA	On request Custom	--
Sesterce France	--	$0.88 /GPU/hr On-demand
Runcrate USA	--	$1.04 /GPU/hr On-demand PCIe
TensorDock Distributed	--	$0.21 /GPU/hr On-demand
Paperspace USA	--	$0.76 /GPU/hr On-demand
Hyperstack UK	--	$0.11 /GPU/hr Reserved
Database Mart USA	--	$0.17 /GPU/hr Reserved
Runpod USA	--	On request Reserved