AMD MI355X vs Nvidia RTX 3090

A data center GPU against a consumer card, one CDNA 4 and one Ampere: 288 GB HBM3e against 24 GB GDDR6X and 8.5x the memory bandwidth on the MI355X.

At a glance

Updated 6 minutes ago

AMD MI355X

Nvidia RTX 3090

Cheapest listed price per GPU per hour by billing type, and the median across providers.

Billing	MI355X	RTX 3090	Difference
On-demand	$8.60	$0.10	86x more expensive on MI355X
Spot	$4.50	$0.07	66x more expensive on MI355X
Reserved	No reserved listings	$0.06
Median	$8.60	$0.25	34x more expensive on MI355X

Median: Median across providers, one vote each, so a large catalog does not outweigh a small one.

Weekly median on-demand price per GPU per hour

Last 12 months

Aggregating historical prices...

Spec	MI355X	RTX 3090	Difference
Memory
Capacity	288 GB HBM3e	24 GB GDDR6X	12x more on MI355X
Bandwidth	8,000 GB/s	936 GB/s	8.5x higher on MI355X
Bus width	8,192-bit	384-bit	21x 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	71.2 TFLOPS	35x higher on MI355X
INT8	5,033.2 TOPS	284.7 TOPS	18x higher on MI355X
FP32	157.3 TFLOPS	35.6 TFLOPS	4.4x higher on MI355X
FP64	78.6 TFLOPS	Not supported	MI355X only
Platform
Interconnect	Infinity Fabric 1,075 GB/s	NVLink 112 GB/s	9.6x higher on MI355X
Board power	1,400 W	350 W	4x higher on MI355X
Released	Q2 2025	Q3 2020	5 years newer on MI355X
Precision
Supported formats	FP4FP6FP8FP16BF16FP32FP64INT8	FP16BF16TF32FP32INT4INT8

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

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