cerebras

Cerebras Systems Inc. is an American artificial intelligence (AI) company with offices in Sunnyvale, San Diego, Toronto, and Bangalore, India. Cerebras builds computer systems for complex AI deep learning applications.

History

Cerebras was founded in 2015 by Andrew Feldman, Gary Lauterbach, Michael James, Sean Lie and Jean-Philippe Fricker. These five founders worked together at SeaMicro, which was started in 2007 by Feldman and Lauterbach and was later sold to AMD in 2012 for $334 million.

In May 2016, Cerebras secured $27 million in series A funding led by Benchmark, Foundation Capital and Eclipse Ventures.

In December 2016, series B funding was led by Coatue Management, followed in January 2017 with series C funding led by VY Capital.

In November 2018, Cerebras closed its series D round with $88 million, making the company a unicorn. Investors in this round included Altimeter, VY Capital, Coatue, Foundation Capital, Benchmark, and Eclipse.

On August 19, 2019, Cerebras announced its first-generation Wafer-Scale Engine (WSE).’

In November 2019, Cerebras closed its series E round with over $270 million for a valuation of $2.4 billion.

In 2020, the company announced an office in Japan and partnership with Tokyo Electron Devices.

In April 2021, Cerebras announced the CS-2 based on the company's Wafer Scale Engine Two (WSE-2), which has 850,000 cores. In August 2021, the company announced its brain-scale technology that can run a neural network with over 120 trillion connections.

In November 2021, Cerebras announced that it had raised an additional $250 million in Series F funding, valuing the company at over $4 billion. The Series F financing round was led by Alpha Wave Ventures and Abu Dhabi Growth Fund (ADG). To date, the company has raised $720 million in financing.

In August 2022, Cerebras was honored by the Computer History Museum in Mountain View, California. The museum added to its permanent collection and unveiled a new display featuring the WSE-2—the biggest computer chip made so far—marking an "epochal" achievement in the history of fabricating transistors as an integrated part.

Cerebras filed its prospectus for initial public offering (IPO) in September 2024, with the intention of listing on the Nasdaq exchange under the ticker 'CBRS'. The prospectus indicated that most of its revenue at the time came from Emirati AI holding company G42. A week after the filing, it was reported that the Committee on Foreign Investment in the United States was reviewing G42's investment into the company, leading to a potential delay in its IPO. In a May 2025 interview, CEO Andrew Feldman said the company had obtained clearance from a U.S. committee to sell shares to G42 and he hopes that Cerebras will go public in 2025.

Cerebras was named to the Forbes AI 50 in April 2024 and the TIME 100 Most Influential Companies list in May 2024.

Technology

The Cerebras Wafer Scale Engine (WSE) is a single, wafer-scale integrated processor that includes compute, memory and interconnect fabric. The WSE-1 powers the Cerebras CS-1, Cerebras’ first-generation AI computer. It is a 19-inch rack-mounted appliance designed for AI training and inference workloads in a datacenter. The CS-1 includes a single WSE primary processor with 400,000 processing cores, as well as twelve 100 Gigabit Ethernet connections to move data in and out. The WSE-1 has 1.2 trillion transistors, 400,000 compute cores and 18 gigabytes of memory.

In April 2021, Cerebras announced the CS-2 AI system based on the 2nd-generation Wafer Scale Engine (WSE-2), manufactured by the 7 nm process of TSMC . It is 26 inches tall and fits in one-third of a standard data center rack.
The Cerebras WSE-2 has 850,000 cores and 2.6 trillion transistors.
The WSE-2 expanded on-chip SRAM to 40 gigabytes, memory bandwidth to 20 petabytes per second and total fabric bandwidth to 220 petabits per second.

In August 2021, the company announced a system which connects multiple integrated circuits (commonly called "chips") into a neural network with many connections.
It enables a single system to support AI models with more than 120 trillion parameters.

In June 2022, Cerebras set a record for the largest AI models ever trained on one device. Cerebras said that for the first time ever, a single CS-2 system with one Cerebras wafer can train models with up to 20 billion parameters. The Cerebras CS-2 system can train multibillion-parameter natural language processing (NLP) models including GPT-3XL 1.3 billion models, as well as GPT-J 6B, GPT-3 13B and GPT-NeoX 20B with reduced software complexity and infrastructure.

In September 2022, Cerebras announced that it can patch its chips together to create what would be the largest-ever computing cluster for AI computing. A Wafer-Scale Cluster can connect up to 192 CS-2 AI systems into a cluster, while a cluster of 16 CS-2 AI systems can create a computing system with 13.6 million cores for natural language processing. The key to the new Cerebras Wafer-Scale Cluster is the exclusive use of data parallelism to train, which is the preferred approach for all AI work.

In November 2022, Cerebras unveiled the supercomputer, Andromeda, which combines 16 WSE-2 chips into one cluster with 13.5 million AI-optimized cores, delivering up to 1 Exaflop of AI computing horsepower, or at least one quintillion (10 to the power of 18) operations per second. The entire system consumes 500 kW, which was a drastically lower amount than somewhat-comparable GPU-accelerated supercomputers.

In November 2022, Cerebras announced its partnership with Cirrascale Cloud Services to provide a flat-rate "pay-per-model" compute time for its ''Cerebras AI Model Studio''. The service is said to reduce the cost—compared to the similar cloud services on the market—by half while increasing speed up to eight times faster.

In July 2023, Cerebras and UAE-based G42 unveiled the world's largest network of nine interlinked supercomputers, Condor Galaxy, for AI model training. The first supercomputer, named Condor Galaxy 1 (CG-1), boasts 4 exaFLOPs of FP16 performance and 54 million cores. In November 2023, the Condor Galaxy 2 (CG-2) was announced, also containing 4 exaFLOPs and 54 million cores.

In March 2024, the companies broke ground on the Condor Galaxy 3 (CG-3), which can hit 8 exaFLOPs of performance and contains 58 million AI-optimized cores.

In March 2024, the company also introduced WSE-3, a 5 nm-based chip hosting 4 trillion transistors and 900,000 AI-optimized cores, the basis of the CS-3 computer. Cerebras also announced a collaboration with Dell Technologies, unveiled in June 2024, for AI compute infrastructure for generative AI.

In August 2024, Cerebras unveiled its AI inference service, claiming to be the fastest in the world and, in many cases, ten to twenty times faster than systems built using the dominant technology, Nvidia's H100 "Hopper" graphics processing unit, or GPU.

As of October 2024, Cerebras' performance advantage for inference is even larger when running the latest Llama 3.2 models. The jump in AI inference performance between August and October is a big one, at a factor of 3.5X, and it opens up the gap between Cerebras CS-3 systems running on premises or in clouds operated by Cerebras.

In March 2025, Cerebras announced six new datacenters across the United States and Europe, increasing the inference capacity twentyfold to over 40 million tokens per second.

In April 2025, Meta announced a partnership with Cerebras to power the new Llama API, offering developers inference speeds up to 18 times faster than with traditional GPU-based solutions. In May, Cerebras announced that it beats NVIDIA Blackwell in Llama 4 Inference with more than double the performance at more than 2,500 tokens per second/user on the 400B parameter Llama 4 Maverick model.

In May 2025, Cerebras unveiled Qwen3-32B, an open-weight LLM model built for smart, high speed reasoning and performance.

Deployments

Customers are reportedly using Cerebras technologies in the hyperscale, pharmaceutical, life sciences, and energy sectors, among others.

CS-1

In 2020, GlaxoSmithKline (GSK) began using the Cerebras CS-1 AI system in their London AI hub, for neural network models to accelerate genetic and genomic research and reduce the time taken in drug discovery. The GSK research team was able to increase the complexity of the encoder models they could generate, while reducing training time. Other pharmaceutical industry customers include AstraZeneca, who was able to reduce training time from two weeks on a cluster of GPUs to two days using the Cerebras CS-1 system. GSK and Cerebras recently co-publishe
research
in December 2021 on epigenomic language models.

Argonne National Laboratory has been using the CS-1 since 2020 in COVID-19 research and cancer tumor research based on the world's largest cancer treatment database. A series of models running on the CS-1 to predict cancer drug response to tumors achieved speed-ups of many hundreds of times on the CS-1 compared to their GPU baselines.

Cerebras and the National Energy Technology Laboratory (NETL) demonstrated record-breaking performance of Cerebras' CS-1 system on a scientific compute workload in November 2020. The CS-1 was 200 times faster than the Joule Supercomputer on the key workload of Computational Fluid Dynamics.

The Lawrence Livermore National Lab’s Lassen supercomputer incorporated the CS-1 in both classified and non-classified areas for physics simulations. The Pittsburgh Supercomputing Center (PSC) has also incorporated the CS-1 in their Neocortex supercomputer for dual HPC and AI workloads. EPCC, the supercomputing center of the University of Edinburgh, has also deployed a CS-1 system for AI-based research.

In August 2021, Cerebras announced a partnership wit
Peptilogics
on the development of AI for peptide therapeutics.

CS-2

In March 2022, Cerebras announced that the Company deployed its CS-2 system in the Houston facilities of TotalEnergies, its first publicly disclosed customer in the energy sector. Cerebras also announced that it has deployed a CS-2 system a
nference
a startup that uses natural language processing to analyze massive amounts of biomedical data. The CS-2 will be used to train transformer models that are designed to process information from piles of unstructured medical data to provide fresh insights to doctors and improve patient recovery and treatment.

In May 2022, Cerebras announced that the National Center for Supercomputing Applications (NCSA) has deployed the Cerebras CS-2 system in their HOLL-I supercomputer. They also announced that the Leibniz Supercomputing Centre (LRZ) in Germany plans to deploy a new supercomputer featuring the CS-2 system along with the HPE Superdome Flex server. The new supercomputing system is expected to be delivered to LRZ this summer. This will be the first CS-2 system deployment in Europe.

In October 2022, it was announced that the U.S. National Nuclear Security Administration would sponsor a study to investigate using Cerebras' CS-2 in nuclear stockpile stewardship computing. The multi-year contract will be executed through Sandia National Laboratories, Lawrence Livermore National Lab, and Los Alamos National Laboratory.

In November 2022, Cerebras and the National Energy Technology Laboratory (NETL) saw record-breaking performance on the scientific compute workload of forming and solving field equations. Cerebras demonstrated that its CS-2 system was as much as 470 times faster than NETL's Joule Supercomputer in field equation modeling.

The 2022 Gordon Bell Special Prize Winner for HPC-Based COVID-19 Research, which honors outstanding research achievement towards the understanding of the COVID-19 pandemic through the use of high-performance computing, used Cerebras' CS-2 system to conduct this award-winning research to transform large language models to analyze COVID-19 variants. The paper was authored by a 34-person team from Argonne National Laboratory, California Institute of Technology, Harvard University, Northern Illinois University, Technical University of Munich, University of Chicago, University of Illinois Chicago, Nvidia, and Cerebras. ANL noted that using the CS-2 Wafer-Scale Engine cluster, the team was able to achieve convergence when training on the full SARS-CoV-2 genomes in less than a day.

Cerebras partnered with Emirati technology group G42 to deploy its AI supercomputers to create chatbots and to analyze genomic and preventive care data. In July 2023, G42 agreed to pay around $100 million to purchase the first of potentially nine supercomputers from Cerebras, each of which capable of 4 exaflops of compute. In August 2023, Cerebras, the Mohamed bin Zayed University of Artificial Intelligence and G42 subsidiary Inception launched Jais, a large language model.

Mayo Clinic announced a collaboration with Cerebras at the 2024 J.P. Morgan Healthcare Conference, offering details on the first foundation model it will develop with the enablement of Cerebras's generative AI computing capability. The solution will combine genomic data with de-identified data from patient records and medical evidence to explore the ability to predict a patient's response to treatments to manage disease and will initially be applied to rheumatoid arthritis. The model could serve as a prototype for similar solutions to support the diagnosis and treatment of other diseases.

At the January 2025 J.P. Morgan Healthcare Conference, Cerebras and Mayo Clinic announced a new genomic foundation model aimed at harnessing the power of advanced AI and HPC to transform genomics, a field rapidly becoming central to personalized healthcare. The new genomic foundation model is designed to improve diagnostics and personalize treatment selection, with an initial focus on Rheumatoid Arthritis.

In May 2024, Cerebras in collaboration with researchers from Sandia National Laboratories, Lawrence Livermore National Laboratory, Los Alamos National Laboratory and the National Nuclear Security Administration, for molecular dynamics simulations in which the team simulated 800,000 atoms interacting with each other, calculating the interactions in increments of one femtosecond at a time. Each step took just microseconds to compute on the Cerebras WSE-2. Although that's still 9 orders of magnitude slower than the actual interactions, it was also 179 times as fast as the Frontier supercomputer. The achievement effectively reduced a year's worth of computation to just two days.

CS-3

In March 2024, Cerebras introduced the CS-3 and third-generation Wafer Scale Engine (WSE-3), which represents the latest development of their technology. It has 2x the performance of CS-2 and hosts 900,000 cores. A CS-3 cluster is capable of training an AI model like Llama2-70B in just one single day. The WSE-3 was recognized by TIME Magazine as a Best Invention of 2024.

In January 2025, Cerebras announced support for DeepSeek's R1 70B reasoning model, running on the latest Cerebras hardware in US-based datacenters.

In April 2025, Cerebras and Canadian photonics company Ranovus announced a contract from DARPA to reduce compute bottleneck challenges.

Cerebras Inference

Cerebras AI Inference services claims to be the fastest in the world and, in many cases, ten to twenty times faster than systems built using the dominant technology, Nvidia's H100 "Hopper" graphics processing unit, or GPU.

In January 2025, Cerebras announced that it would support DeepSeek's R1 70B reasoning model at 1,600 tokens/second, which the company claims is 57x faster than any R1 provider using GPUs.

In February 2025, Cerebras and Mistral AI announced a partnership and helped the French AI player achieve a speed record. Mistral released an app called Le Chat that it said can respond to user questions with 1,000 words per second. Cerebras said it is providing the computer power behind those results.

Also in February 2025, Cerebras announced a partnership with Perplexity AI that promises to deliver near-instantaneous AI-powered search results at speeds previously thought impossible. The collaboration centers on Perplexity's new Sonar model which runs on Cerebras chips at 1,200 tokens per second making it one of the fastest AI search systems available. Cerebras appears to be leveraging this momentum to establish itself as the go-to provider for high-speed AI inference.

See also

* Wafer-scale integration
* Wafer-level packaging
* Semiconductor device fabrication
* Transistor count
* Deep learning processor

References

External links

* {{Official website, https://cerebras.net/

Cerebras' presentation at Hot Chips 34 (2022)

Computer companies of the United States
Companies based in California
Companies based in Sunnyvale, California
Companies based in Silicon Valley
Computer hardware companies
Semiconductor companies of the United States
Fabless semiconductor companies
Computer companies established in 2016
2016 establishments in California