For leading price performance in the cloud and on-premises

With Habana’s first-generation Intel® Gaudi® AI deep learning processor, customers benefit from the most cost-effective, high-performance training and inference alternative to comparable GPUs. This is the deep learning architecture that enables AWS to deliver up to 40% better price/performance training with its Intel Gaudi AI accelerator-based DL1 instances—as compared to comparable Nvidia GPU-based instances. Intel Gaudi’s AI accelerator efficient architecture also enables Supermicro to provide customers with equally significant price performance advantage over GPU-based servers with the Supermicro X12 Intel Gaudi AI Training Server.
AWS Logo

Intel Gaudi AI accelerator in the Cloud

Get started with Intel Gaudi AI – based Amazon EC2 DL1 Instances

Supermicro logo

Intel Gaudi AI accelerator in the Data Center

Build Intel Gaudi AI processors into
your data center with Supermicro

Quote mark icon
“When benchmarking training YOLOv5 on COCO, we found Habana Intel Gaudi AI accelerator HPUs to outperform the incumbent NVIDIA A100 GPUs by $0.25 per epoch. That’s 25% more epochs per dollar.”

What makes Intel Gaudi AI accelerator
so efficient?

gaudi-diagram
16nm
Process technology
DL Optimized
Matrix multiplication engine
8 Programmable
Tensor processor cores
32 GB
Onboard HBM2
24 SRAM
10 Integrated
100G Ethernet ports
“At Leidos, we rank our solutions to our customers using the parameters of speed, scale, security, and usability. Our solution on Amazon EC2 DL1 Instances checks all the boxes.”
-Chetan Paul, VP of Technology and Innovation Federal Health

Massive and flexible system scaling with Intel Gaudi AI accelerator

Every first-generation Intel Gaudi AI processor integrates ten 100 Gigabit Ethernet ports of RDMA over Converged Ethernet (RoCE2) on chip to deliver unmatched scalability, enabling customers to efficiently scale AI training from one processor to 1000s to nimbly address expansive compute requirements of today’s deep learning workloads.

Get details in this video >

Options for building
Intel Gaudi AI accelerator systems on premises

For customers who want to build out on-premises systems, we recommend the Supermicro X12 Intel Gaudi AI Training Server, which features eight Intel Gaudi AI processors. For customers who wish to configure their own Intel Gaudi AI-based servers, we provide reference model options, the Intel Gaudi AI HLS-1 and HLS-1H.
For more information on these server options, please see more details >

Scalability icon

Making developing on Intel Gaudi AI accelerator Fast and Easy: SynapseAI® Software Suite

Optimized for deep learning model development and to ease migration of existing GPU-based models to Intel Gaudi AI platform hardware. It integrates PyTorch and TensorFlow frameworks and supports a rapidly growing array of computer vision, natural language processing and multi-modal models. In fact, over 50K models on Hugging Face are easily enabled on Gaudi with the Habana Optimum software library.
Getting started with model migration is as easy as adding 2 lines of code, and for expert users who wish to program their own kernels, Habana offers the full tool-kit and libraries to do that as well. SynapseAI software supports training and inference of models on first-gen Intel Gaudi AI and Intel Gaudi2 AI .
For more information on the how Habana is making it easy to migrate existing or building new models on Gaudi, see our SynapseAI product page >

SynapseAI icon
“Our benchmark showed that the Habana-based Amazon EC2 DL1 training instance delivered from 51% to 73% better price performance compared to the current GPU-based AWS EC2 instances on our deep learning benchmarking training tasks.”
– www.aldwin.eu/livres-blancs/benchmark-habana-labs/

Learn more about
Intel Gaudi AI accelerators.

Contact us