Home
Compare GPUs
MI250 vs L4

AMD MI250 vs Nvidia L4

Two data center GPUs, one CDNA2 and one Ada Lovelace: 128 GB HBM2e against 24 GB GDDR6, 11x the memory bandwidth on the MI250, and Infinity Fabric 400 GB/s on the MI250 where the L4 is PCIe only.

At a glance

Memory

MI250: 128 GB
L4: 24 GB

5.3x more on MI250

Memory bandwidth

MI250: 3,200 GB/s
L4: 300 GB/s

11x higher on MI250

Cheapest on-demand

MI250: --
L4: $0.44 /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	L4
On-demand	No on-demand listings	$0.90 /GPU/hr from $0.44
Reserved	-- /GPU/hr from $1.30 (12 mo)	$0.53 /GPU/hr from $0.32 (36 mo)
Spot	No spot listings	-- /GPU/hr from $0.34

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 spot on the L4: 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 L4 price history

Weekly median on-demand price per GPU per hour

Last 12 months

Aggregating historical prices...

Specs comparison

Spec	MI250	L4	Difference
Memory
Capacity	128 GB HBM2e	24 GB GDDR6	5.3x more on MI250
Bandwidth	3,200 GB/s	300 GB/s	11x higher on MI250
Bus width	8,192-bit	192-bit	43x wider on MI250
Compute · dense
FP8	Not supported	242 TFLOPS	L4 only
FP16 / BF16	362.1 TFLOPS	121 TFLOPS	3x higher on MI250
INT8	362.1 TOPS	242 TOPS	1.5x higher on MI250
FP32	90.5 TFLOPS	30.3 TFLOPS	3x higher on MI250
FP64	45.3 TFLOPS	Not supported	MI250 only
Platform
Interconnect	Infinity Fabric 400 GB/s	PCIe only	Infinity Fabric on MI250 only
Board power	500 W	72 W	6.9x higher on MI250
Released	Q4 2021	Q1 2023	16 months newer on L4
Precision
Supported formats	FP16BF16FP32FP64INT4INT8	FP8FP16BF16TF32FP32INT4INT8

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

Source: MI250 datasheet and L4 datasheet.

Where to rent them

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

Provider	MI250 /GPU/hr	L4 /GPU/hr
Cirrascale	Reserved (12mo) $1.30	--
GPU.ai	--	On-Demand $0.44
Runpod	--	On-Demand $0.49
UpCloud	--	On-Demand $0.69
Scaleway	--	On-Demand $0.91
Sesterce	--	On-Demand $1.04
Runcrate	--	On-Demand $1.05
Hexabyte	--	On-Demand $0.42
Koyeb	--	On-Demand $0.70