Home
Compare GPUs
MI210 vs L40

AMD MI210 vs Nvidia L40

Two data center GPUs, one CDNA2 and one Ada Lovelace: 64 GB HBM2e against 48 GB GDDR6, 90% more memory bandwidth on the MI210, and Infinity Fabric 300 GB/s on the MI210 where the L40 is PCIe only.

At a glance

Memory

MI210: 64 GB
L40: 48 GB

1.3x more on MI210

Memory bandwidth

MI210: 1,638 GB/s
L40: 864 GB/s

1.9x higher on MI210

Cheapest on-demand

MI210: --
L40: $0.45 /GPU/hr

MI210 is not sold on demand

Price comparison

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

Billing	MI210	L40
On-demand	No on-demand listings	$1.00 /GPU/hr from $0.45
Reserved	-- /GPU/hr from $0.86 (1 mo)	-- /GPU/hr from $0.70
Spot	No spot listings	$0.78 /GPU/hr from $0.27

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 MI210 and reserved on the L40: 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

MI210 vs L40 price history

Weekly median on-demand price per GPU per hour

Last 12 months

Aggregating historical prices...

Specs comparison

Spec	MI210	L40	Difference
Memory
Capacity	64 GB HBM2e	48 GB GDDR6	1.3x more on MI210
Bandwidth	1,638 GB/s	864 GB/s	1.9x higher on MI210
Bus width	4,096-bit	384-bit	11x wider on MI210
Compute · dense
FP8	Not supported	362 TFLOPS	L40 only
FP16 / BF16	181 TFLOPS	181.1 TFLOPS	Within 5%
INT8	181 TOPS	362 TOPS	2x higher on L40
FP32	45.3 TFLOPS	90.5 TFLOPS	2x higher on L40
FP64	22.6 TFLOPS	Not supported	MI210 only
Platform
Interconnect	Infinity Fabric 300 GB/s	PCIe only	Infinity Fabric on MI210 only
Board power	300 W	300 W	Identical
Released	Q1 2022	Q4 2022	7 months newer on L40
Precision
Supported formats	FP16BF16FP32FP64INT4INT8	FP8FP16BF16TF32FP32INT4INT8

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

Source: MI210 datasheet and L40 datasheet.

Where to rent them

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

Provider	MI210 /GPU/hr	L40 /GPU/hr
Leaseweb	Reserved (1mo) $0.86	--
Vast.ai	--	On-Demand $0.45
GPU.ai	--	On-Demand $0.69
Runpod	--	On-Demand $0.69
Thunder Compute	--	On-Demand $0.79
Massed Compute	--	On-Demand $0.86
Sesterce	--	On-Demand $0.97
Runcrate	--	On-Demand $0.97
Oblivus	--	On-Demand $1.05