Deep learning is taking the next leap forward. Increasingly sophisticated and complex data, models, and techniques will allow AI to move beyond identifying information to understanding context – enabling a level of “common sense” for reasoning and decision-making. But this evolution is pushing existing compute to a breaking point.
Intel® Neural Network Processors (Intel NNPs) were built from the ground up to break through existing memory and data flow bottlenecks, enabling distributed learning algorithms and systems that will scale up deep learning reasoning, using more advanced forms of AI to go beyond the conversion of data into information-turning data into global knowledge.
Next-gen deep learning training systems must efficiently scale out for larger, more complex models. The Intel® Nervana™ Neural Network Processor-T (NNP-T) was designed for this purpose, able to train even the most advanced deep learning models at near-linear scaling and incredible efficiency. Its careful balance of compute, memory and networking ensures maximum utilization at low time-to-train in real-world training scenarios, whether using small clusters or large PODs, and even at smaller matrix sizes.
Enterprise-scale AI deployments are rapidly increasing inference cycle volume for an increasingly diverse set of applications with varying latency and power needs. The Intel® Nervana™ Neural Network Processor-I (NNP-I) is built for ultra-efficient multi-modal inferencing of intense, near-real-time, high-volume compute. It can accommodate exponentially larger, more complex models running up to dozens of models and networks in parallel. Focused on power- and budget-efficiency, it offers 10W-75W form factors for real-world scaling and flexibility.
Related Resources
Blog Post
11.19.19
Artificial intelligence is speeding up industries, research, and commerce around the globe. For enterprises, it’s not about if they’ll incorporate…
Video
11.12.19
The Intel Nervana Neural Network Processor (NNP) is a purpose built architecture for deep learning. This new architecture provides the…
Video
11.12.19
Dr. Naveen Rao reveals the industry tailwinds pushing Intel’s diverse AI hardware and software portfolio into the spotlight, featuring new…
Blog Post
11.12.19
It's an exciting day for Intel and the AI community we work to enable. Today, we're proud to provide significant…
Blog Post
11.06.19
At Intel, our vision of the future is one in which AI is infused everywhere, helping to make the world…
Blog Post
10.10.19
In my over 20 years of working with Intel, I’ve learned something very important: moving an industry forward is not…
Blog Post
09.18.19
Inference is where the transformative power of AI meets real-world applications. It’s how a smart speaker recognizes a plethora of…
Blog Post
09.17.19
The recent advancements in artificial intelligence has been driven by abundant compute, large data sets, and the re-emergence of deep…
For further details and notices, please visit www.intel.ai/benchmarks