AMD MI250 vs Nvidia H800

The MI250 has 60% more memory and 60% more bandwidth.

At a glance

Memory

60% more on MI250

MI250: 128 GB

H800: 80 GB

Memory bandwidth

60% higher on MI250

MI250: 3.20 TB/s

H800: 2.00 TB/s

Cheapest on-demand

MI250 is not sold on demand

MI250: --

H800: --

Price comparison

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

Billing	MI250	H800	Difference
Reserved	-- /GPU/hr from $1.30 (12 mo)	No reserved 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 on the MI250: 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

MI250 vs H800 price history

Weekly median on-demand price per GPU per hour

Last 12 months

Aggregating historical prices...

Specs comparison

Spec	MI250	H800	Difference
Memory
Capacity	128 GB HBM2e	80 GB HBM2e	60% more on MI250
Bandwidth	3,200 GB/s	2,000 GB/s	60% higher on MI250
Bus width	8,192-bit	5,120-bit	60% wider on MI250
Compute · dense
FP8	Not supported	1,513 TFLOPS	H800 only
FP16 / BF16	362.1 TFLOPS	756.5 TFLOPS	2.1x higher on H800
INT8	362.1 TOPS	1,513 TOPS	4.2x higher on H800
FP32	90.5 TFLOPS	51 TFLOPS	77% higher on MI250
FP64	45.3 TFLOPS
Platform
Interconnect	Infinity Fabric 400 GB/s	NVLink 400 GB/s	Identical
Board power	500 W	350 W	43% higher on MI250
Released	Q4 2021	Q1 2023	16 months newer on H800
Precision
Supported formats	FP16BF16FP32FP64INT4INT8	FP8FP16BF16TF32FP32FP64INT8