Home
Compare GPUs
MI250 vs GH200

AMD MI250 vs Nvidia GH200

Two data center GPUs, one CDNA2 and one Grace Hopper: 128 GB HBM2e against 96 GB HBM3 and 25% more memory bandwidth on the GH200.

At a glance

Memory

MI250: 128 GB
GH200: 96 GB

1.3x more on MI250

Memory bandwidth

MI250: 3.20 TB/s
GH200: 4.00 TB/s

1.2x higher on GH200

Cheapest on-demand

MI250: --
GH200: $6.50 /GPU/hr

MI250 is not sold on demand

Price comparison

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

Billing	MI250	GH200
On-demand	No on-demand listings	-- /GPU/hr from $6.50
Reserved	-- /GPU/hr from $1.30 (12 mo)	No reserved listings
Spot	No spot listings	Sold out

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 and on-demand 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

MI250 vs GH200 price history

Weekly median on-demand price per GPU per hour

Last 12 months

Aggregating historical prices...

Specs comparison

Spec	MI250	GH200	Difference
Memory
Capacity	128 GB HBM2e	96 GB HBM3	1.3x more on MI250
Bandwidth	3,200 GB/s	4,000 GB/s	1.2x higher on GH200
Bus width	8,192-bit
Compute · dense
FP8	Not supported	1,979 TFLOPS	GH200 only
FP16 / BF16	362.1 TFLOPS	989.5 TFLOPS	2.7x higher on GH200
INT8	362.1 TOPS	1,979 TOPS	5.5x higher on GH200
FP32	90.5 TFLOPS	67 TFLOPS	1.4x higher on MI250
FP64	45.3 TFLOPS	34 TFLOPS	1.3x higher on MI250
Platform
Interconnect	Infinity Fabric 400 GB/s	NVLink 900 GB/s	2.2x higher on GH200
Released	Q4 2021	Q2 2023	2 years newer on GH200
Precision
Supported formats	FP16BF16FP32FP64INT4INT8	FP8FP16BF16TF32FP32FP64INT8