AMD MI350X vs Nvidia GB200

Two data center GPUs, one CDNA 4 and one Grace Blackwell: 288 GB HBM3e against 186 GB HBM3e.

At a glance

Memory

MI350X: 288 GB
GB200: 186 GB

1.5x more on MI350X

Memory bandwidth

MI350X: 8.00 TB/s
GB200: 8.00 TB/s

Identical

Cheapest on-demand

MI350X: --
GB200: $10.50 /GPU/hr

MI350X is not sold on demand

Price comparison

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

Billing	MI350X	GB200
On-demand	No on-demand listings	-- /GPU/hr from $10.50
Reserved	-- /GPU/hr from $4.76 (12 mo)	-- /GPU/hr from $10.58
Spot	-- /GPU/hr from $4.00	-- /GPU/hr from $27.04

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 and spot on the MI350X and on-demand, reserved and spot on the GB200: 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

MI350X vs GB200 price history

Weekly median on-demand price per GPU per hour

Last 12 months

Aggregating historical prices...

Specs comparison

Spec	MI350X	GB200	Difference
Memory
Capacity	288 GB HBM3e	186 GB HBM3e	1.5x more on MI350X
Bandwidth	8,000 GB/s	8,000 GB/s	Identical
Bus width	8,192-bit
Compute · dense
FP4	9,227.5 TFLOPS	10,000 TFLOPS	1.1x higher on GB200
FP8	4,614 TFLOPS	5,000 TFLOPS	1.1x higher on GB200
FP16 / BF16	2,309.6 TFLOPS	2,500 TFLOPS	1.1x higher on GB200
INT8	4,614 TOPS	5,000 TOPS	1.1x higher on GB200
FP32	144.2 TFLOPS	80 TFLOPS	1.8x higher on MI350X
FP64	72.1 TFLOPS	40 TFLOPS	1.8x higher on MI350X
Platform
Interconnect	Infinity Fabric 1,075 GB/s	NVLink 1,800 GB/s	1.7x higher on GB200
Board power	1,000 W	1,200 W	1.2x higher on GB200
Released	Q2 2025	Q1 2024	15 months newer on MI350X
Precision
Supported formats	FP4FP6FP8FP16BF16FP32FP64INT8	FP4FP6FP8FP16BF16TF32FP32FP64INT8

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

Source: MI350X datasheet and GB200 datasheet.

Where to rent them

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

Provider	MI350X /GPU/hr	GB200 /GPU/hr
DigitalOcean	Reserved (12mo) $4.76	--
CoreWeave	--	On-Demand $10.50
Azure	--	On-Demand $27.04
AWS	--	Reservation $10.58
CUDO	--	Custom on request
Canopy Wave	--	Custom on request
Nebius	--	Custom on request
Together AI	--	Custom on request