In the News

The next decade might demonstrate waferscale as one of only a few bridges across the post-Moore’s Law divide, at least for some applications. Luckily for the only maker of such a system, Cerebras, those areas are among the most high-value workloads in large industry.

As the Only European Provider of Cerebras Cloud, Green AI Cloud Delivers AI Super Compute in an Easy-to-Use Solution for AI and ML Applications, Data Science and Simulation Workloads

Cerebras is not your typical AI chip company. And that's a good thing. We already have a lot of knock-offs taking on NVIDIA, and they all seem to do a good job of knocking only themselves out. Now Cerebras has announced new cloud access on Cirrascale with a fixed-price schedule to train foundation models on its hardware.

With Predictable Fixed Pricing, Faster Time to Solution, and Unprecedented Flexibility and Ease of Use, Customers Can Train GPU-Impossible Sequence Lengths and Keep Trained Weights

As one of the first customers to leverage Cerebras’ new Andromeda AI supercomputer to design next set of customer-specific models, Jasper expects to vastly improve the accuracy of generative AI models

Cerebras unveiled its new AI supercomputer Andromeda at SC22. With 13.5 million cores across 16 Cerebras CS-2 systems, Andromeda boasts an exaflop of AI compute and 120 petaflops of dense compute. Its computing workhorse is Cerebras’ wafer-scale, manycore processor, WSE-2.

ACM, the Association for Computing Machinery, awarded the 2022 ACM Gordon Bell Special Prize for High Performance Computing-Based COVID-19 Research to an international team for their project “GenSLMs: Genome-Scale Language Models Reveal SARS-CoV-2 Evolutionary Dynamics.” The Prize recognizes outstanding research achievement toward the understanding of the COVID-19 pandemic through the use of high-performance computing.

HPCwire, the leading publication for news and information for the high performance computing industry, announced the winners of the 2022 HPCwire Readers’ and Editors’ Choice Awards at the Supercomputing Conference (SC22) taking place this week in Dallas, Texas. Tom Tabor, CEO of Tabor Communications, unveiled the list of 137 winners across 21 categories just before the opening gala reception.

The research leveraged several major supercomputers, as well as a major AI-focused system from Cerebras Systems, applying LLMs to transform how new and emergent variants of viruses like SARS-CoV-2 are identified and classified. (Subsequently, this work won the Gordon Bell Special Prize for HPC-based COVID-19 research.)

We live in a world that has become accustomed to cloud computing, and now it is perfectly acceptable to do timesharing on such machines to test ideas out. This is precisely what Cerebras is doing as it stands up a 13.5 million core AI supercomputer nicknamed “Andromeda” in a colocation facility run by Colovore in Santa Clara, the very heart of Silicon Valley.

Exascale Cerebras Andromeda cluster packs more cores than 1,954 Nvidia A100 GPUs.

Silicon Valley startup Cerebras Systems, known in the industry for its dinner plate-sized chip made for artificial intelligence work, on Monday unveiled its AI supercomputer called Andromeda, which is now available for commercial and academic research.

Delivering more than 1 Exaflop of AI compute and 120 Petaflops of dense compute, Andromeda is one of the largest AI supercomputers ever built, and is dead simple to use

The Cerebras CS-2 and its Wafer-Scale Engine Outperforms Leading Supercomputer by Over Two Orders of Magnitude in Time to Solution, Delivering CPU and GPU-Impossible Performance

Thermonuclear warheads, so hot right now

Investigating stockpile stewardship applications for world’s largest computer chip

Beyond the Wafer Scale Engine, their challenge doesn't end with just making the biggest chip ever

Comparison to Cerebras

Beyond the Wafer Scale Engine, their challenge doesn't end with just making the biggest chip ever

A conversation with Andrew Feldman, co-founder and CEO of Cerebras Systems.

The hottest trend in AI is the emergence of massive models such as Open AI’s GPT-3. These models are surprising even its developers with capabilities that some claim approach human sentience. Our analyst, Alberto Romero, shows the capabilities of these models in the image below on the right, which he created with a simple prompt using Midjourney, an AI similar to Open AI’s DALL·E. Large Language Models will continue to evolve to become powerful tools in businesses from Pharmaceuticals to Finance, but first they need to become easier and more cost-effective to create.

Chipmaker Cerebras is patching its chips – already considered the world’s largest – to create what could be the largest-ever computing cluster for AI computing.

The Cerebras Wafer-Scale Cluster delivers near-perfect linear scaling across hundreds of millions of AI-optimized compute cores while avoiding the pain of the distributed compute.

Cerebras, the company that released a gigantic (850,000 core) processor, claims their chip will democratize the hardware needed to train and run very large language models by eliminating the need to distribute computation across thousands of smaller GPUs.

Cerebras' Wafer Scale Engine beeindruckt nicht nur mit der Größe, sondern auch mit der Architektur. Der Chip berechnet 7,5 Petaflops.

Cerebras Systems said that its CS-2 high-performance computing accelerator is now capable of training artificial intelligence models on 20X longer sequences than "traditonal" hardware, including graphics processing units. The company expects this new capability to lead to advances in natural language processing in drug discovery, particularly by allowing researchers to examine individual genes in the context of many thousands of surrounding genes.

Jean-luc Chatelain, Applied Intelligence CTO, talks with Andrew Feldman, Founder and CEO of Cerebras Systems about transformers, massive models and the future of AI. They discuss how we are moving from a world of a large amount of models to fewer more powerful models known as transformers. Hear what they think this means for the future of AI.

Our 106th episode with a summary and discussion of last week's big AI news! Apologies for a lack of consistent episode releases lately, we've been really busy...

It’s that time of year again—Hot Chips will soon be upon us. Taking place as a virtual event on August 21–23, the conference will once again present the very latest in microprocessor architectures and system innovations.

In the professional side, it’s very easy to underestimate the power of communication. In my field, when you interact with executives, customers, users…you need to be able to translate and present your ideas and the key points to any audience. It’s very valuable for people — especially those of us in the technical field or mathematics — to understand and execute the power of communication.

It’s that time of year again—Hot Chips will soon be upon us. Taking place as a virtual event on August 21–23, the conference will once again present the very latest in microprocessor architectures and system innovations.

Artificial intelligence is being deployed across a number of new applications, from improving performance and reducing power in a wide range of end devices to spotting irregularities in data movement for security reasons.

We talk a lot about chips on this channel, so it stands to reason that there's a conference about the hottest chips around! Introducing Hot Chips 2022, this year's place-to-be about silicon.

　人工知能（AI）スタートアップのCerebras Systemsは米国時間8月3日、米カリフォルニア州マウンテンビューにあるコンピューター歴史博物館で行われたセレモニーで、その技術の伝統を担っていることを認められた。同博物館は、Cerebrasが開発した史上最大のコンピューターチップである同社の第2世代AIチップ、「Wafer-Scale Engine 2（WSE-2）」の展示を開始した。このチップは2021年に発表されたもので、同社の新型スーパーコンピューター「CS-2」に搭載されている。

Dans le secteur de la technologie, chaque invention s'appuie sur les succès et les échecs qui l'ont précédée.

Début août, la start-up d'IA Cerebras Systems a été récompensée pour avoir perpétué cette tradition lors d'une cérémonie au Computer History Museum de Mountain View, en Californie.

How machine learning could help save threatened species from extinction, new algorithm aces university math course questions, a 175B parameter publicly available chatbot

Der wohl größte Computerchip der Welt könnte bald ein künstliches Gehirn simulieren, das quantitativ an das menschliche heranreicht. Warum es dennoch kaum wie ein Mensch denken wird.

The problem is that the people closest to the technology — the people explaining it to the public — live with one foot in the future. They sometimes see what they believe will happen as much as they see what is happening now.
“There are lots of dudes in our industry who struggle to tell the difference between science fiction and real life,” said Andrew Feldman, chief executive and founder of Cerebras, a company building massive computer chips that can help accelerate the progress of A.I.

To get a better sense of why VCs are betting on chip startups, Insider compiled a list of several startups investors say are bringing new technological strides to the field. The firms range from chips designed for artificial intelligence and machine learning to semiconductors embedded in clothes. The firms on this list were picked in consult with VCs and analysts. They were chosen for their funding, leadership, and other criteria to help determine which will stick around.

The silicon chip, or integrated circuit (IC), is one of humankind’s most magnificent, complex, and transformative creations.

When Cerebras Systems had its coming out at Hot Chips in August 2019, the hardware community wasn’t sure what to think. Attendees were understandably skeptical of the novel “wafer-scale” technology, not to mention an estimated power envelope of ~15 kilowatts for the chip alone. In the intervening three years, the company – under the direction of founder and CEO Andrew Feldman – has won over early critics through a series of impressive milestones.

Company honored by Computer History Museum for cracking the code of making giant chips, with 'stunning' implications.

Am 14. Juli hat das Leibniz-Rechenzentrum (LRZ) der Bayerischen Akademie der Wissenschaften sein 60jähriges Bestehen gefeiert. Das Hochleistungsrechenzentrum, das unter anderem den Supercomputer „SuperMUC“ beherbergt ist der IT-Dienstleister der Münchner Universitäten und bayerischen Hochschulen sowie Kooperationspartner wissenschaftlicher Einrichtungen in Bayern, Deutschland und Europa. Doch Alter hat in der IT nur bedingt Wert. Wie geht es weiter?

To continue making our gadgets more powerful, engineers have worked out a new way to get around the barriers to making microchips faster: Just make them bigger.

There is growing interest in democratizing large language models and making them available to a broader audience. However, the hardware barriers of large language models have gone mostly unaddressed. This is one of the problems that Cerebras, a startup that specializes in AI hardware, aims to solve with its Wafer-Scale processor. In an interview with TechTalks, Cerebras CEO Andrew Feldman discussed the hardware challenges of LLMs and his company’s vision to reduce the costs and complexity of training and running large neural networks.

Researchers at the U.S. National Energy Technology Laboratory (NETL) are helping the National Center for Atmospheric Research (NCAR) unlock the potential of an advanced artificial intelligence (AI) computing resource to perform critical climate modeling that could lead to better climate change predictions.

Two Cerebras Systems engineers are finalists for VentureBeat's Women in AI awards!

Discussion of last week's big AI news -- including Cerebras Systems Sets Record for Largest AI Models Ever Trained on a Single Device!

US semiconductor startup Cerebras claims that it has trained the largest AI model on a single device. The company trained AI models with 20 billion parameters on its Wafer Scale Engine 2 (WSE-2) chip, the world's largest chip.

Cerebras has shown off the capabilities of its second–generation wafer–scale engine, announcing it has set the record for the largest AI model ever trained on a single device.

For the first time, a natural language processing network with 20 billion parameters, GPT–NeoX 20B, was trained on a single device. Here’s why that matters.

Cerebras said it can reduce the engineering time to run large NLP models from months to minutes, making it more cost-effective and accessible.

In my first video talking about AI Hardware on this channel, I talk in detail about an exciting Cerebras CS-2 announcement!

Democratizing large AI Models without HPC scaling requirements.

Cerebras Systems makes the largest chip in the world, but is already thinking about its upcoming AI chips as learning models continue to grow at breakneck speed.

Cerebras Systems said it has set the record for the largest AI models ever trained on a single device, which in this case is a giant silicon wafer with hundreds of thousands of cores.

Chip startup Cerebras has developed a foot-wide piece of silicon, compared to average chips measured in millimeters, that makes training AI cheap, and easy.

Cerebras engineer Vitaliy Chiley joins Machine Learning Street Talk to discuss deep learning, sparsity and different compute architectures for artificial intelligence.

At ISC 2022 I had the opportunity to sit down with Andy Hock at Cerebras. This was very cool after I told Cerebras and HPE that STH would not cover the HPE Superdome plus one Cerebras CS-2 being installed at LRZ for ISC, Andy still agreed to meet. As a result, we get a little article about the experience. This was one of the more interesting conversations since we veered fairly far from the Cerebras product. Instead, what I wanted to do was just give a perspective, reinforced by this discussion, on what Cerebras is doing to differentiate itself against a huge competitor (NVIDIA.)

Die Anforderungen an die Hardware steigen in KI und beim Hochleistungsrechnen stetig. Die Antwort darauf sind neue Ansätze bei Chip- und Rechnerdesign. Eine neuartige Kombi-Architektur wird jetzt das Leibniz-Rechenzentrum (LRZ) in Garching implementieren.

Cerebras Systems, the pioneer in high performance artificial intelligence (AI) computing, today announced that the National Center for Supercomputing Applications (NCSA) has deployed the Cerebras CS-2 system in their HOLL-I supercomputer.

Cerebras Systems, the pioneer in high performance artificial intelligence (AI) computing, shared news about their many supercomputing partners including European Parallel Computing Center (EPCC), Leibniz Supercomputing Centre (LRZ), Lawrence Livermore National Laboratory, Argonne National Laboratory (ANL), the National Center for Supercomputing Applications (NCSA), and the Pittsburgh Supercomputing Center (PSC).

A European supercomputing hub near Munich, called the Leibniz Supercomputing Centre, is deploying Cerebras Systems’ CS-2 AI system as part of an internal initiative called Future Computing to assess alternative computing technologies to inject more speed into the region’s scientific research.

Wafer madness hits the LRZ in HPE Superdome supercomputer wrapper

Als erster Standort in Europa hat das Leibniz-Rechenzentrum (LRZ) ein CS-2-System mit Cerebras' WSE-2 gekauft, welches effizient und schnell ist.

The Leibniz Supercomputing Centre (LRZ) in Germany plans to deploy a new supercomputer featuring the HPE Superdome Flex server and the Cerebras CS-2 system.

Hewlett Packard Enterprise (HPE) has announced it is building a powerful new AI supercomputer in collaboration with Cerebras Systems, maker of the world’s largest chip.

The Cerebras CS-2 system was awarded a Best of Show award as an innovative solution to important problems facing the life sciences industry.

As the pandemic swept across the world, virtually every research supercomputer lit up to support Covid-19 investigations. But even as the world transformed, the fairly stable status quo of simulation-based scientific computing was itself beginning to more rapidly change with the burgeoning field of AI and AI-specific accelerators, including the Cerebras CS-2.

When the chip is the size of a big pizza pie... that's Cerebras

Leading research institutes are choosing HPE Superdome Flex as the basis of new supercomputing systems designed to accelerate AI. Both the University of Edinburgh and Pittsburgh Supercomputing Center (PSC) are combining Superdome Flex with Cerebras CS-1, an AI accelerator based on the largest processor in the industry.

2,6 Billionen (!) Transistoren, 850.000 Rechenkerne, 40 GByte On-Chip-Memory, 20 kW Leistungsaufnahme: Die Wafer Scale Engine 2 von Cerebras ist in jeder Hinsicht extrem. Dabei ist der Name Programm: Der 7-nm-Chip belegt nahezu einen kompletten 300-mm-Wafer. Mittlerweile ist der Chip am LRZ in Garching bei München im Einsatz.

Cerebras—in collaboration with French multinational TotalEnergies—has announced the development of a massively scalable stencil algorithm: a development made possible by the use of one of Cerebras’ CS-2 systems.

Cerebras' CSoft software update to version 1.2 gives developers using PyTorch and TensorFlow access to their powerful CS-2 systems through Cerebras Cloud.

Cerebras Systems, makers of the world’s largest chip, has announced that its CS-2 system now supports PyTorch and TensorFlow which will make it possible for researchers to quickly and easily train models with billions of parameters.

AI hardware and software vendor Cerebras Systems released an updated version of its platform that includes integrated support for the open source TensorFlow and PyTorch machine learning frameworks.

The Cerebras CS-2 is the world’s fastest AI system and is powered by its Wafer-Scale Engine 2 (WSE-2) CPU. With the release of version 1.2 of the Cerebras Software Platform (CSoft), the CS-2 now supports additional machine learning (opens in new tab) frameworks which will give developers even more choice when it comes to the types of models they want to run.

When I first saw an image of the 'wafer-scale engine' from AI hardware startup Cerebras, my mind rejected it. The company's current product is about the size of an iPad and uses 2.6 trillion transistors that contribute to over 850,000 cores.

Good news for those who like their AI chips big: Cerebras Systems has expanded support for the popular open-source PyTorch and TensorFlow machine-learning frameworks on the Wafer-Scale Engine 2 processors that power its CS-2 system.

Innovators don’t see limitations – they see challenges. And that’s exactly what happened when Andrew Feldman and his team at Cerebras Systems were told that it was impossible to build a computer chip that could deliver the same performance as hundreds of graphics processing units.

Investors are funneling billions of dollars into startups like Cerebras Systems that make chips designed for artificial-intelligence applications, which have largely avoided the supply-chain constraints and backlogs faced by larger chip makers, startup executives, investors and industry analysts say.

The ability to harness vast amounts of health data using advanced AI technology will lead to new discoveries and insights needed to improve patient care

High performance AI compute company Cerebras Systems and nference, an AI-driven health technology company, today announced a collaboration to accelerate natural language processing (NLP) for biomedical research and development by orders of magnitude with a Cerebras CS-2 system installed at the nference headquarters in Cambridge, Mass.

Venture capitalists funneled billions into semiconductor startups like Cerebras Systems in 2021, we're told, targeting designers of machine-learning technologies that fulfill specific or niche needs.

Cerebras Systems, a pioneer in high performance artificial intelligence (AI) compute, and nference, an AI-driven health technology company, today announced a collaboration to accelerate natural language processing (NLP) for biomedical research and development by orders of magnitude with a Cerebras CS-2 system installed at the nference headquarters in Cambridge, Mass.

The Cerebras CS-2 system for AI is a piece of art

Cerebras should be ready to focus on customer success stories in 2022, having built an accessible, if not expensive, system-level solution that has been adopted in HPC and a few enterprises.

As 2022 begins, the progress and continuing evolution of AI is in full swing across the worlds of industry, manufacturing, retail, banking and finance, healthcare and medicine and an expanding range of other fields.

Natural language processing (NLP) has been one of the hottest sectors in AI over the past two years. Will the string of big data breakthroughs continue into 2022? We checked in with industry experts -- including Natalia Vassilieva from Cerebras Systems -- to find out.

Wafer-scale AI accelerator company Cerebras Systems has devised a memory bottleneck solution at the far end of the scale. At Hot Chips, the company announced MemoryX, a memory extension system for its CS-2 AI accelerator system aimed at high-performance computing and scientific workloads.

Cerebras Systems said its Wafer Scale Engine 2 chip is the “largest AI processor ever made,” consisting of 2.6 trillion transistors, 850,000 cores and 40 GB of on-chip memory. The startup said those specifications give the WSE-2 chip a massive advantage over GPU competitors.

Last April, Cerebras Systems topped its original, history-making AI processor with a version built using a more advanced chipmaking technology. The result was a more than doubling of the on-chip memory to an impressive 40 gigabytes, an increase in the number of processor cores from the previous 400,000 to a speech-stopping 850,000, and a mind-boggling boost of 1.4 trillion additional transistors.

One company contributing to the fight with artificial intelligence and machine learning is Cerebras Systems. The second iteration of the company’s Wafer-Scale Engine machine, the CS-2 system, plays an important role as part of the AI Testbed at Argonne National Laboratory and has contributed to a multi-agency COVID-19 reproduction study that was nominated as a Gordon Bell Special Prize finalist.

Lawrence Livermore National Laboratory has established the AI Innovation Incubator (AI3), a collaborative hub aimed at uniting experts in artificial intelligence from LLNL, industry and academia including Cerebras Systems to advance AI for large-scale scientific and commercial applications.

EE Times interviews Cerebras CEO Andrew Feldman on leadership, personal projects and the biggest threat to society posed by technology today

It’s a bird, it’s a plane, it’s the largest AI processor ever made! In this week’s Fish Fry podcast, Andy Hock (Cerebras Systems) joins me to chat about the largest AI processor ever made – the 7 nm wafer scale engine 2, the details of their brain-scale AI training, and how Cerebras Systems is democratizing access to high performance AI computation.

It’s a bird, it’s a plane, it’s the largest AI processor ever made!  In this week’s Fish Fry podcast, Andy Hock (Cerebras Systems) joins me to chat about the largest AI processor ever made –  the 7 nm wafer scale engine 2, the details of their brain-scale AI training, and how Cerebras Systems is democratizing access to high performance AI computation.

Cerebras Systems VP Andy Hock predicts that, "In 2022, AI will continue to grow as a valuable and critical workload for enterprise organizations across industries. We will see a larger number of teams investing in world-class AI computing to accelerate their research and business..."

Simon McIntosh-Smith discusses the role of the ExCALIBUR project in ensuring that UK research is at the forefront of HPC -- including a Cerebras CS-1

This year has been marked by continued uncertainty and upheaval of our consumer habits and supply-and-demand balance. The dynamic business landscape has been challenging for even the most successful corporations, let alone for startups that are earlier in their efforts to establish market traction and growth.

As part of their series about the women leading the Artificial Intelligence industry, Tyler Gallagher interviewed Natalia Vassilieva, Director of Product, Machine Learning at Cerebras Systems.

GlaxoSmithKline Plc hopes to double its drug success rate to 20% by using AI, teaming up with partners including DNA testing provider 23andMe Holding Co. and Cerebras Systems Inc., an upstart chipmaker that provides computer systems to crunch very large data sets.

Additional innovations such as three-dimensional chip stacking and Cerebras wafer-scale chips may help extend Moore’s law for another decade or two, at least.

Cerebras Systems and G42 signed a memorandum of understanding (MOU) at GMIS, committing to offer high-performance AI capabilities to the Middle East. G42 will enhance its technology stack with Cerebras’ CS-2 systems to give unrivaled AI compute capabilities to its partners and the broader ecosystem.

Artificial intelligence (AI) compute solutions provider Cerebras Systems and G42, the UAE-based AI and cloud computing company, have signed a memorandum of understanding (MOU) at GMIS, under which they will bring high performance AI capabilities to the Middle East.

Semiconductor startups like Cerebras Systems are offering novel ways to improve the performance, efficiency, economics and bandwidth of servers and beyond with new takes on CPUs, accelerators and connectivity solutions.

DUBAI: It seems only fitting that the world’s biggest computer chip, developed by US-based Cerebras, be deployed in the Middle East, home to some of the world’s entertaining superlatives — biggest tower, deepest swimming pool, among many other “firsts.”

Cerebras Systems and G42 today announced at GMIS the signing of a memorandum of understanding (MOU) under which they will bring high performance AI capabilities to the Middle East. G42, who manages the region’s largest cloud computing infrastructure, will upgrade its technology stack with Cerebras’ industry-leading CS-2 systems to deliver unparalleled AI compute capabilities to its partners and the broader ecosystem.

Eye On A.I.

Creator of the world’s fastest AI supercomputer among top-ranked companies in the Artificial Intelligence category

Businesswire