Home
Compare GPUs
Gaudi2 vs A100

Intel Gaudi2 vs Nvidia A100

Two data center GPUs, one Gaudi (2nd Gen) and one Ampere: 96 GB HBM2e against 80 GB HBM2e and 27% more memory bandwidth on the Gaudi2.

At a glance

Memory

Gaudi2: 96 GB
A100: 80 GB

1.2x more on Gaudi2

Memory bandwidth

Gaudi2: 2.45 TB/s
A100: 1.94 TB/s

1.3x higher on Gaudi2

Cheapest on-demand

Gaudi2: $1.03 /GPU/hr
A100: $0.80 /GPU/hr

22% cheaper on A100

Price comparison

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

Billing	Gaudi2	A100	Difference
On-demand	$1.21 /GPU/hr from $1.03	$1.77 /GPU/hr from $0.80	32% cheaper on Gaudi2
Reserved	-- /GPU/hr from $0.58 (12 mo)	$1.66 /GPU/hr from $0.65 (3 mo)
Spot	No spot listings	$1.20 /GPU/hr from $0.68

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 Gaudi2: 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

Gaudi2 vs A100 price history

Weekly median on-demand price per GPU per hour

Last 12 months

Aggregating historical prices...

Specs comparison

Spec	Gaudi2	A100	Difference
Memory
Capacity	96 GB HBM2e	80 GB HBM2e 40 GB on SXM, PCIe	1.2x more on Gaudi2
Bandwidth	2,450 GB/s	1,935 GB/s 2,039 GB/s on SXM 1,555 GB/s on SXM, PCIe	1.3x higher on Gaudi2
Bus width		5,120-bit
Compute · dense
FP8	865 TFLOPS	Not supported	Gaudi2 only
FP16 / BF16	432 TFLOPS	312 TFLOPS	1.4x higher on Gaudi2
INT8		624 TOPS
FP32		19.5 TFLOPS
FP64	Not supported	9.7 TFLOPS	A100 only
Platform
Interconnect	RoCE 600 GB/s	NVLink 600 GB/s	Identical
Board power	600 W	300 W	2x higher on Gaudi2
Released	Q2 2022	Q2 2020	2 years newer on Gaudi2
Precision
Supported formats	FP8FP16BF16TF32FP32INT8	FP16BF16TF32FP32FP64INT4INT8

Figures are for each model's headline configuration. Where a form factor carries a different number, it is named under the figure.

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

Source: Gaudi2 datasheet and A100 datasheet.

Where to rent them

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

Provider	Gaudi2 /GPU/hr	A100 /GPU/hr
Runcrate	On-Demand $1.21	--
Sesterce	On-Demand $1.21	--
Cyfuture AI	On-Demand $1.03	--
Vast.ai	--	On-Demand $0.80
GPU.ai	--	On-Demand $0.89
Thunder Compute	--	On-Demand $1.09
Lium	--	On-Demand $1.23
Massed Compute	--	On-Demand $1.35
Runpod	--	On-Demand $1.39
Spheron	--	On-Demand $1.43
Packet·ai	--	On-Demand $1.43