Neural Magic FulFills On The Promise Of Software-Delivered AI With The Latest MLPerf™ Inference v3.0 Results


Share post:

Neural Magic, AI leader in sparse inferencing technology, today announced a 6X improvement over their groundbreaking results from their last round submission with open engineering consortium, MLCommons®. Neural Magic’s latest results in MLPerf™ Inference v3.0, validate that customers can achieve high-performance AI using just x86 CPU architectures.

“We are so excited about our latest performance results, validated by MLPerf Inference v3.0,” said Brian Stevens, Chief Executive Officer, Neural Magic. “By applying our specialized compound sparsity algorithms with our patented DeepSparse inference runtime, we were able to achieve a boost in CPU performance by 1,000X while reducing power consumption by 92% over other inference solutions. The numbers prove users can have GPU speeds to support AI projects with just software and off-the-shelf processors.”

This year, Neural Magic used 4th Gen AMD EPYC™ processors for their benchmark testing. Neural Magic’s software stack takes advantage of continued innovations in AMD EPYC processors, such as AVX-512 and VNNI instructions as well as advanced features like highly performant DDR5 memory and a core count up to 96 cores, to unlock new possibilities for delivering better than GPU speeds on x86 CPUs.

Also Read: Role of Open Source Software in Driving Innovation and Efficiency in Digital Enterprise

“Neural Magic’s MLPerf™ Inference v3.0 results, benchmarked using our latest 4th Gen AMD EPYC processors, prove customers can achieve outstanding levels of AI inference performance for deep learning projects on x86 based CPUs, ” said Kumaran Siva, Corporate Vice President, Strategic Business Development, AMD.

With Neural Magic, data scientists can achieve breakthrough performance with their deep learning models, while lowering computational expenses and simplifying operations. Read this blog for more details on Neural Magic’s MLPerf Inference v3.0 results.

TalkDev Bureau
TalkDev Bureau
The TalkDev Bureau has five well-trained writers and journalists, well versed in B2B enterprise technology industry, and constantly in touch with industry leaders for the latest trends, opinions, and other inputs- to bring you the best and latest in the domain.


Please enter your comment!
Please enter your name here


Related articles

Radix IoT Mango 5 Optimizes Unparalleled Intuitive IoT Scalability for Mission-Critical Monitoring

Radix IoT, LLC today announced the release of Mango 5, advancing large-scale IoT multi-site deployments and monitoring scalability to unprecedented heights....

Persistent expands relationship with AWS to adopt Amazon CodeWhisperer

Persistent Systems, a global Digital Engineering leader, is strengthening its relationship with Amazon Web Services (AWS) and becoming...

KDDI Deploys DriveNets Network Cloud

DriveNets, a provider of cloud-native networking solutions, declared that KDDI corporation has successfully deployed DriveNets Network Cloud as...

Voxel51 Introduces VoxelGPT

Voxel51, a provider of data-centric computer vision and machine learning software, introduces, VoxelGPT. It is an extension of FiftyOne,...