Kalray Joins MLCommons as Founding Member

Kalray Joins MLCommons as Founding Member
EN
Kalray Joins MLCommons as Founding Member Download
242.56 KB
FR
Kalray participe à la création de MLCommons, consortium dédié à l'Intelligence Artificielle, en tant que membre fondateur Download
436.18 KB

MLCommons Unites 30+ Global Technology and Academic Leaders
in Artificial Intelligence and Machine Learning


Grenoble – France, December 3, 2020 –
Kalray (Euronext Growth Paris: FR0010722819 – ALKAL), a pioneer in processors for new intelligent systems, today announced it has joined MLCommons as founding member. MLCommons is a not for profit, engineering consortium founded by 30+ global technology and academic leaders in AI and Machine Learning to accelerate machine learning innovation and broaden access to this critical technology for the public good.

Initially formed as MLPerf, MLCommons boasts a founding board that includes representatives from Alibaba, Facebook AI, Google, Intel, NVIDIA and Professor Vijay Janapa Reddi of Harvard University with the objective to deliver industry-wide benchmarks, best practices and datasets to speed computer vision, natural language processing, and speech recognition development for all.

MLCommons has a clear mission – accelerate Machine Learning innovation to ‘raise all boats’ and increase its positive impact on society” said Peter Mattson, President of MLCommons. “We are excited to build on and extend the scope of the already impressive impact of MLPerf, by bringing together our global partners across industry and academia to develop technologies that benefit everyone.”

Whereas most of the solutions available on the market today are pure AI accelerators, Kalray’s MPPA® (“Massively Parallel Processor Array“) intelligent processor has been designed to build complete intelligent systems that will be at the heart of a growing number of applications such as autonomous vehicles, intelligent storage servers, next generation data centers, 5G Telecom Edge and more. Kalray intelligent processors are not only capable of executing AI algorithms but can also simultaneously execute a wide set of different processing such as mathematical algorithms, signal processing, network or storage software stacks. This heterogenous multi-processing capability is a critical requirement for Edge Computing markets and is a unique capability of Coolidge™, 3rd generation of MPPA® processor available now.

Eric Baissus, President and CEO of Kalray, has declared: “We are delighted to join MLCommons, alongside prestigious founding members at this important time in the growth and adoption of AI and Machine Learning. AI is one critical component of next generation intelligent systems. Our goal will be to work with the other founding members to define and benchmark the AI use cases that will be important for next generation of AI-based applications, especially in data centers, automotive and edge computing.”

To ease access to MPPA®, Kalray has developed Kalray Neural Network 3.0 (KaNN™), a software platform for AI application development. KaNN™ allows developers to seamlessly port their AI-based algorithms from wellknown machine learning frameworks including Caffe®, Torch® and TensorFlow® onto Kalray’s MPPA® intelligent processor.

Together with Kalray, the semiconductor sector is well represented with several founding member companies including AMD, ARM, Cerebras Systems, Centaur Technology, d-Matrix, FuriosaAI, Grai Matter Labs, Graphcore, Groq, Horizon Robotics, Intel, MediaTek, Neuchips, Nvidia, Qualcomm, Samsung Electronics, Shanghai Enflame Technology, Syntiant, Tenstorrent, Xilinx.