Coreweave S-1 Filing: How the AI Cloud Provider Plans to Compete and Grow

Coreweave's S-1 filing reveals ambitious plans to compete with major cloud providers. The company aims to achieve this through its patented serverless architecture, which allows for more efficient use of resources.

Coreweave's technology is designed to be highly scalable, making it an attractive option for businesses with large computing needs. By leveraging this scalability, Coreweave plans to reduce costs and increase profit margins.

The company's focus on serverless architecture also enables it to provide a more flexible and secure cloud experience. This is particularly important for businesses that require high levels of data protection and compliance.

Coreweave's growth strategy involves expanding its customer base and increasing its market share through strategic partnerships and acquisitions.

A fresh viewpoint: Do Businesses Prefer Cash or Credit

CoreWeave's business model is built around providing high-performance cloud computing services tailored for AI and other GPU-intensive workloads. The company operates 32 data centers running over 250,000 GPUs, supported by 360 MW of active power.

CoreWeave generates revenue primarily through usage fees and enterprise contracts, with a focus on delivering large-scale GPU capacity to its customers. Its usage-based model allows customers to scale their GPU usage up or down as needed.

The company's partnership with Nvidia has been a key factor in its growth, with Nvidia investing in CoreWeave and supplying it with hardware. CoreWeave was among the first cloud providers to deploy Nvidia's H100 and H200 GPUs upon release.

Related reading: How Does Your Business Innovate or Add Value for Customers

CoreWeave's transformation from crypto mining to AI infrastructure is a remarkable story of pivoting to success.

The company was founded in 2017 by three former commodities traders: Michael Intrator, Brian Venturo, and Brannin McBee, under the name Atlantic Crypto.

They initially focused on mining Ethereum, but in 2019, they made a strategic decision to pivot towards AI infrastructure.

This pivot proved timely, as the release of OpenAI's ChatGPT in 2022 sparked increased demand for AI computing resources.

CEO Michael Intrator controls approximately 38% of voting power, which is a significant concentration of leadership.

The company's founders identified a gap in the cloud provider market, where traditional providers offered limited compute options at high prices.

This change allowed CoreWeave to tap into the growing demand for AI computing resources and establish itself as a provider of infrastructure optimized specifically for AI computing requirements.

CoreWeave's managed Kubernetes service and various abstraction layers designed for AI workloads have been key to its success.

The company's leadership has not publicly responded to concerns about GPU oversupply, but it's clear that their focus on AI infrastructure has been a crucial factor in their growth.

Intriguing read: Coreweave Leadership

CoreWeave generates revenue through a usage-based model, charging customers based on the capacity and usage of its GPU clusters. This means customers only pay for the compute time they use, making it a cost-effective option for demanding workloads.

The company's focus is on AI and GPU-intensive workloads, which is reflected in its proprietary software, including the SUNK orchestration system and Tensorizer optimization tools. These tools are designed to maximize GPU utilization across clients.

CoreWeave competes with major cloud providers like Amazon AWS and Microsoft Azure, but its market positioning is that of a specialized alternative. Its data centers are purpose-built for AI, providing an edge in efficiency and performance.

The company has developed a close partnership with Nvidia, which supplies it with hardware and has invested in CoreWeave. This partnership has allowed CoreWeave to rapidly deliver large-scale GPU capacity.

As of December 31, 2024, CoreWeave operates 32 data centers running more than 250,000 GPUs, supported by 360 MW of active power.

Additional reading: Gpu Mining 2024

CoreWeave targets two key customer segments: AI Enterprises and AI Natives. This approach acknowledges the diverse adoption patterns emerging in the rapidly evolving AI ecosystem.

CoreWeave's customer base is dominated by Microsoft, which accounted for 62% of total revenues in 2024. The company's filing notes that two customers generated 77% of revenues that year.

The emergence of large language models and Gen AI applications has driven unprecedented demand for specialized AI computing resources since late 2022. Organizations developing and deploying these AI models require purpose-built computing environments prioritizing performance, scalability, and cost efficiency.

CoreWeave operates in a competitive market for AI infrastructure, facing competition from hyperscale cloud providers like AWS, Google Cloud, and Microsoft Azure, which are expanding their AI-specific offerings and leveraging their extensive global infrastructure.

The company also faces competition from specialized AI infrastructure startups, such as Lambda Labs, Crusoe, RunPod, Together AI, and Vultr, which are all privately held and closely watching CoreWeave's progress.

CoreWeave's competitive advantage lies in its purpose-built infrastructure specifically optimized for AI applications, potentially offering performance, cost, or flexibility advantages compared to general-purpose cloud infrastructure.

The constrained supply of advanced GPUs and other specialized AI accelerators creates both challenges and opportunities in this market, with companies having established supply chains and purchasing relationships holding significant advantages in accessing critical hardware components during periods of supply limitation.

Recommended read: Mncs Advantages

CoreWeave's revenue skyrocketed from $16 million in 2022 to $229 million in 2023, then jumped to $1.92 billion in 2024, representing 737% year-over-year growth.

The company's rapid expansion is largely driven by the booming demand for its AI-focused cloud services. This growth is rare, and what was a relatively small business just two years ago has evolved into a nearly $2 billion revenue company.

CoreWeave's biggest customer is Microsoft, which accounted for 35% of total revenue in 2023 and grew to 62% in 2024. Another major client, possibly an AI firm, also contributes significantly to the company's income.

In 2024, the top two customers contributed 77% of total revenue, highlighting the company's reliance on a few large clients. This is evident in the Master Services Agreement signed with OpenAI in March 2025, which secures a major long-term contract.

CoreWeave agreed to create a special purpose vehicle (SPV) to manage the infrastructure supporting OpenAI's workloads, with agreements being finalized as of March 11, 2025. To fulfill this deal, the company will issue OpenAI $350 million worth of common stock at its IPO price.

Despite its surging revenue, CoreWeave is not yet profitable, as it continues to invest heavily in expansion. Net losses widened from $593.7 million in 2023 to $863.4 million in 2024, driven by massive infrastructure investments, equipment depreciation, and interest expenses on debt.

Here's an interesting read: Jarrow March

CoreWeave is targeting a massive $2.3 billion in debt and equity financing, including a reported $400 million investment from Nvidia, its principal GPU supplier.

The company has already secured significant funding from Magnetar Capital, fueling its aggressive expansion to meet surging demand from AI-focused customers.

CoreWeave's IPO filing is led by Morgan Stanley, J.P. Morgan, and Goldman Sachs, joined by a broader group of underwriters.

The company has established a dual-class stock structure, with Class A shares carrying one vote each and Class B shares carrying ten votes each, primarily held by founders and early investors.

This governance structure will maintain concentrated control in the hands of CoreWeave's founders after the IPO, with CEO Michael Intrator, CSO Brian Venturo, and CDO Brannin McBee collectively retaining majority voting control.

CoreWeave had cash and cash equivalents of $1.36 billion at the end of 2024 and a total debt of $7.92 billion.

The company's revenue showed significant year-over-year growth, increasing from $16 million in 2022 to $229 million in 2023 and to $1.9 billion in 2024.

You might like: Trade-off Theory of Capital Structure

CoreWeave is a specialized AI cloud provider that's taking a different approach to cloud computing. They're one of the first to pursue public market funding.

Their infrastructure is optimized specifically for AI computing requirements, which is a significant departure from the hyperscale cloud providers dominating the market.

CoreWeave offers bare metal computing resources that remove virtualization layers to maximize performance. This is a key differentiator in the AI infrastructure market.

The company also provides a managed Kubernetes service and various abstraction layers designed for AI workloads. This helps developers focus on their applications rather than the underlying infrastructure.