Intel Gaudi 2 vs Nvidia A4000

The Gaudi 2 is newer, with 6x the memory and 5.5x the bandwidth. The A4000 is typically cheaper.

At a glance

Memory

6x more on Gaudi 2

Gaudi 2: 96 GB

A4000: 16 GB

Memory bandwidth

5.5x higher on Gaudi 2

Gaudi 2: 2,450 GB/s

A4000: 448 GB/s

Cheapest on-demand

80% cheaper on A4000

Gaudi 2: $1.04 /GPU/hr

A4000: $0.21 /GPU/hr

Price comparison

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

Billing	Gaudi 2	A4000	Difference
On-demand	$1.10 /GPU/hr from $1.04	$0.76 /GPU/hr from $0.21	31% cheaper on A4000
Reserved	-- /GPU/hr from $0.59 (12 mo)	$0.25 /GPU/hr from $0.11

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 Gaudi 2: 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

Gaudi 2 vs A4000 price history

Weekly median on-demand price per GPU per hour

Last 12 months

Aggregating historical prices...

Specs comparison

Spec	Gaudi 2	A4000	Difference
Memory
Capacity	96 GB HBM2e	16 GB GDDR6	6x more on Gaudi 2
Bandwidth	2,450 GB/s	448 GB/s	5.5x higher on Gaudi 2
Bus width		256-bit
Compute · dense
FP8	865 TFLOPS	Not supported	Gaudi 2 only
FP16 / BF16	432 TFLOPS	76.7 TFLOPS	5.6x higher on Gaudi 2
FP32		19.2 TFLOPS
Platform
Interconnect	RoCE 600 GB/s	PCIe only	RoCE on Gaudi 2 only
Board power	600 W	140 W	4.3x higher on Gaudi 2
Released	Q2 2022	Q2 2021	13 months newer on Gaudi 2
Precision
Supported formats	FP8FP16BF16TF32FP32INT8	FP16BF16TF32FP32INT4INT8

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

Source: Gaudi 2 datasheet and A4000 datasheet.

Where to rent them

Providers that rent these GPUs, with each one's cheapest rate. Providers that rent both are listed first.

Provider	Gaudi 2	A4000
Runcrate USA	$1.43 /GPU/hr On-demand PCIe	$1.04 /GPU/hr On-demand PCIe
Sesterce France	$1.21 /GPU/hr On-demand	$0.88 /GPU/hr On-demand
Cyfuture AI India	$1.04 /GPU/hr On-demand	--
Denvr Canada	$1.10 /GPU/hr On-demand	--
Runpod USA	--	$0.25 /GPU/hr On-demand
TensorDock Distributed	--	$0.21 /GPU/hr On-demand
Paperspace USA	--	$0.76 /GPU/hr On-demand
Hyperstack UK	--	$0.11 /GPU/hr Reserved
Database Mart USA	--	$0.17 /GPU/hr Reserved
Cirrascale USA	--	$0.28 /GPU/hr Reserved