Cerebras Systems raises $1B at $23B valuation as Benchmark doubles down with $225M investment

Cerebras Systems Secures Massive Funding Round

Cerebras Systems announced it raised $1 billion in a Series H funding round that valued the AI chipmaker at $23 billion, representing a nearly threefold increase from the $8.1 billion Cerebras Systems valuation reached just four months earlier in September 1

1

2

2

. The late-stage funding was led by Tiger Global, with participation from AMD, Benchmark Capital, Fidelity Management and Research Company, Atreides Management, Alpha Wave Global, Altimeter, Coatue, and 1789 Capital 3

3

. This marks the AI chipmaker's second billion-dollar raise since September, underscoring the intense investor interest in companies positioned to challenge established players in the AI chip supply chain.

Benchmark Capital Creates Special Funds for Investment

A significant portion of the new capital came from Benchmark Capital, one of Cerebras' earliest backers that led the startup's $27 million Series A in 2016 1

1

. The prominent Silicon Valley firm invested at least $225 million in this latest funding round, according to sources familiar with the deal. Because Benchmark deliberately keeps its funds under $450 million, the firm raised two separate vehicles, both called 'Benchmark Infrastructure,' specifically to fund the Cerebras investment 1

1

. This unusual move demonstrates the firm's conviction in Cerebras' potential to disrupt the AI hardware market dominated by Nvidia.

Wafer-Scale Engine Technology Sets Cerebras Apart

What distinguishes this Nvidia rival is the physical scale of its processors. The company's Wafer Scale Engine, with its flagship WSE-3 AI chip announced in 2024, measures approximately 8.5 inches on each side and packs 4 trillion transistors into a single piece of silicon—19 times more than Nvidia's Blackwell B200 graphics card 1

1

4

4

. The chip is manufactured from nearly an entire 300-millimeter silicon wafer, whereas traditional chips are thumbnail-sized fragments cut from these wafers. This architecture delivers 900,000 specialized cores working in parallel, allowing the system to process AI calculations without shuffling data between multiple separate chips—a major bottleneck in conventional GPU clusters 1

1

4

4

.

OpenAI Partnership Validates AI Infrastructure Approach

The funding comes as Cerebras gains momentum in the AI infrastructure race. Last month, the Sunnyvale, California-based company signed a multi-year agreement worth more than $10 billion to provide 750 megawatts of computing power to OpenAI through 2028 1

1

3

3

. The partnership aims to help OpenAI deliver faster response times for complex AI queries. OpenAI CEO Sam Altman, who is also an early investor in Cerebras, has been seeking alternatives to Nvidia for AI inference chips as the company looks to diversify its chip supplies 2

2

. About half of the WSE-3 processor's surface area is dedicated to hosting a 44-gigabyte pool of SRAM memory, enabling it to run many AI models without moving data to off-chip HBM memory, which speeds up AI model training and inference tasks by more than 20 times compared to competing systems 1

1

4

4

.

Path to IPO Complicated by G42 Relationship

Cerebras' journey to going public has faced obstacles related to its relationship with G42, a UAE-based AI firm that accounted for 87% of Cerebras' revenue as of the first half of 2024 1

1

. G42's historical ties to Chinese technology companies triggered a national security review by the Committee on Foreign Investment in the United States, delaying Cerebras' initial IPO plans and prompting the company to withdraw an earlier filing in early 2025. By late last year, G42 had been removed from Cerebras' investor list, clearing the way for a fresh IPO attempt 1

1

. The company filed for an IPO in September 2024, disclosing that it generated $136.4 million in revenue during the first six months of 2024, a more than tenfold increase from the same period a year earlier, while losses narrowed to $66.6 million from $77.8 million 4

4

. Cerebras is now preparing for a public debut in the second quarter of 2026, according to Reuters 1

1

.

Competition Intensifies in Data Centers Market

The race to develop data centers for supporting AI technology has made the chips required for computing power a prized commodity, fueling Nvidia's rise to become the most valuable company in the world 2

2

. AI-linked companies have continued to draw billions in private financing as corporations and governments race to scale the technology 5

5

. The fundraising underscores a broader trend of companies staying private for longer with abundant capital available outside of public markets 2

2

. According to Cerebras, customers can link together 2,048 CS-3 appliances—water-cooled systems that ship the WSE-3 chip—into a cluster with 256 exaflops of aggregate computing capacity, enough to train a large language model with 24 trillion parameters 4

4

. This capability positions Cerebras as a key player alongside Nvidia and AMD in the evolving AI chip supply chain, particularly as major AI players seek to diversify their hardware suppliers.