Nvidia A10 vs Nvidia H20

The H20 is newer, with 4x the memory and 6.7x the bandwidth.

At a glance

Memory

4x more on H20

A10: 24 GB

H20: 96 GB

Memory bandwidth

6.7x higher on H20

A10: 600 GB/s

H20: 4,000 GB/s

Cheapest on-demand

A10: $1.00 /GPU/hr

H20: --

Price comparison

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

Billing	A10	H20
On-demand	$1.55 /GPU/hr from $1.00	No on-demand listings
Reserved	-- /GPU/hr from $0.71 (12 mo)	No reserved listings
Spot	-- /GPU/hr from $0.59	No spot 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 and spot on the A10: 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

A10 vs H20 price history

Weekly median on-demand price per GPU per hour

Last 12 months

Aggregating historical prices...

Specs comparison

Spec	A10	H20	Difference
Memory
Capacity	24 GB GDDR6	96 GB HBM3	4x more on H20
Bandwidth	600 GB/s	4,000 GB/s	6.7x higher on H20
Bus width	384-bit
Platform
Interconnect	PCIe only	PCIe only
Board power	150 W	400 W	2.7x higher on H20
Released	Q2 2021	Q1 2024	3 years newer on H20
Precision
Supported formats	FP16BF16TF32FP32INT4INT8	FP8FP16BF16TF32FP32FP64INT8

Source: A10 datasheet .

Where to rent them

Providers that rent these GPUs, with each one's cheapest rate. No provider we track offers both right now.

Provider	A10	H20
Lambda Labs USA	$1.29 /GPU/hr On-demand PCIe	--
Sesterce France	$1.42 /GPU/hr On-demand	--
Runcrate USA	$1.68 /GPU/hr On-demand PCIe	--
OVH France	$1.00 /GPU/hr On-demand	--
Cerebrium USA	$1.10 /GPU/hr On-demand	--
Alibaba Cloud China	$1.27 /GPU/hr On-demand	--
Oracle Cloud USA	$2.00 /GPU/hr On-demand	--
Azure USA	$3.20 /GPU/hr On-demand	--