Prospectus Summary
PROSPECTUS SUMMARY This summary highlights selected information that is presented in greater detail elsewhere in this prospectus. This summary does not contain all of the information you should consider before investing in our Class A common stock. This summary contains forward-looking statements that involve risks and uncertainties. You should carefully read this prospectus in its entirety before investing in our Class A common stock, including the sections titled “Risk Factors,” “Management’s Discussion and Analysis of Financial Condition and Results of Operations,” “Special Note Regarding Forward-Looking Statements,” and our consolidated financial statements and the accompanying notes included elsewhere in this prospectus before making an investment decision.Overview
CoreWeave powers the creation and delivery of the intelligence that drives innovation.
We are the AI HyperscalerTM driving the AI revolution.1 Our CoreWeave Cloud Platform consists of our proprietary software and cloud services that deliver the software and software intelligence needed to manage complex AI infrastructure at scale. Our platform supports the development and use of ground-breaking models and the delivery of the next generation of AI applications that are changing the way we live and work across the globe—our platform is trusted by some of the world’s leading AI labs and AI enterprises, including Cohere Inc. (“Cohere”), International Business Machines Corporation (“IBM”), Meta Platforms, Inc. (“Meta”), Microsoft Corporation (“Microsoft”), Mistral AI SAS (“Mistral”), and NVIDIA Corporation (“NVIDIA”).
We believe AI is the next frontier for innovation in technology, driving productivity and efficiency gains and enabling new business models in nearly every industry and organization. According to IDC, AI will generate a cumulative global economic impact of $20 trillion, or 3.5% of global GDP, by 2030. The generalized cloud infrastructure that drove the cloud revolution beginning in the 2000s was built to host websites, databases, and SaaS apps that have fundamentally different needs than the high performance requirements of AI. As workloads and technologies evolve, so too must the infrastructure and cloud software and services that power them. We believe we are at the start of a new cloud era that will drive the AI revolution. The opportunity for a purpose-built AI cloud platform, including the infrastructure and integrated software, is massive. Based on market research from Bloomberg Intelligence, total spending on AI inference/fine-tuning, AI workload monitoring, and training infrastructure, including AI servers, AI storage, training compute, cloud workloads, and networking, will reach approximately $399 billion by 2028. For AI to reach its full potential, it needs a purpose-built AI cloud platform with infrastructure and managed cloud services that are delivered in an efficient, automated, and highly performant way. Enter CoreWeave, the AI HyperscalerTM.
We purpose-built our CoreWeave Cloud Platform to be the infrastructure and application platform for AI. Our platform manages the complexity of engineering, assembling, running, and monitoring state-of-the-art infrastructure at a massive scale to deliver high performance and efficiency to AI workloads. Through our proprietary software capabilities, we enable our customers to achieve substantially higher total system performance and more favorable uptime relative to other AI offerings within existing infrastructure cloud environments and unlock speed at scale. By delivering more compute cycles to AI workloads and thereby reducing the time required to train models, our capabilities can significantly accelerate the time to solution for customers in the ongoing hyper-competitive race to build the next bleeding-edge AI models. For example, in June 2023, our NVIDIA H100 Tensor Core GPU training cluster completed the MLPerf benchmark test (which
benchmarks how quickly a system can train a model from scratch) in eleven minutes—a record and 29 times faster than the next best competitor at the time of the benchmark test.
These efficiencies also extend from training to inference use cases, as our CoreWeave Cloud Platform significantly improves both run-time efficiency for inference workloads and enables overall higher AI application uptime. These performance gains help to ensure lower performance-adjusted costs and a superior end-user experience. The supercomputers we build to power our platform are optimized to support many types of AI workloads, and they are augmented by our suite of cloud services to deliver meaningful time and cost savings to customers through our orchestration, automation, and monitoring capabilities.
Our multidisciplinary, customer-centric team has a proven ability to conceptualize, design, and implement solutions to solve the most complex engineering challenges in the pursuit of furthering AI. We hire individuals who help contribute to and maintain a culture centered around solving the most complex AI infrastructure scaling, performance, and reliability challenges.
Customers utilize our platform through a set of cloud services comprising Infrastructure Services, Managed Software Services, and Application Software Services, all augmented by our Mission Control and Observability software. Our comprehensive and integrated cloud services work together as a suite to deliver state-of-the-art compute, networking, and storage. These services enable the provisioning of infrastructure, the orchestration of workloads, and the proactive monitoring of our customers’ training and inference environments to increase performance and minimize interruptions.
Our CoreWeave Cloud Platform is hosted in our distributed network of active purpose-built data centers that are interconnected using low latency connections to major metropolitan areas, and incorporate state-of-the-art data center networking equipment, enhanced access to power and, where appropriate, the latest liquid cooling technologies. As of December 31, 2024, our 32 data centers were running more than 250,000 GPUs in total, and were supported by more than 360 MW of active power. Our total contracted power extends to approximately 1.3 GW as of December 31, 2024, which we expect to roll out over the coming years.
We benefit from robust collaborations with leading chipmakers, original equipment manufacturers (“OEMs”), and software providers to supply us with infrastructure components and other products. We have a proven track record of rapidly expanding our power capacity to support the growth of our data center footprint along with our collection of managed cloud services. We deploy a sophisticated financing strategy and have efficiently financed the development of additional compute capacity through the use of asset-backed debt, having raised total commitments of $12.9 billion in debt through December 31, 2024 to support the development of our platform.
Our customers include some of the world’s leading AI labs and AI enterprises—the builders and integrators of AI—who depend on our platform for their core products and most promising innovations. We deliver significant benefits to our customers in terms of overall performance, time to market, and reduced cost of ownership, which results in our customers making large, long-term initial commitments and expanding those commitments with us over time. The vast majority of our revenue today is from multi-year committed contracts, whereby a customer purchases access to our platform over the contract term on a take-or-pay basis. We also sell access to our platform on an on-demand basis through a pay-as-you-go model. As of December 31, 2024, we had $15.1 billion of remaining performance obligations reflecting an increase of 53%, from $9.9 billion as of December 31, 2023.
Our ability to abstract away the complexity our customers would face in assembling, managing, and deploying this infrastructure themselves establishes us as a critical partner and leads to long-term, durable relationships that have the potential to expand over time. As evidence of this, three of our top five committed contract customers by total contract value (“TCV”) as of December 31, 2024 signed agreements for additional capacity within 12 months of their respective initial purchase dates. These agreements, measured during each respective 12-month period from the initial date of signing, represent a cumulative increase of approximately $7.8 billion in committed spend and a multiple of approximately 4x on initial contract value. Our deep relationships with customers are a competitive advantage, and our first-to-market track record with highly performant technology gives customers confidence in choosing CoreWeave.
Some of the most ambitious compute-intensive projects in the world are powered by our platform, and our business has grown rapidly since our inception. Our revenue was $16 million, $229 million, and $1.9 billion for the years ended December 31, 2022, 2023, and 2024, respectively, representing year-over-year growth of 1,346% and 737%, respectively. During these periods, we continued to invest in growing our business to capitalize on our market opportunity. As a result, our net loss for the years ended December 31, 2022, 2023, and 2024 was $31 million, $594 million, and $863 million, respectively. Our adjusted net loss for the years ended December 31, 2022, 2023, and 2024 was $27 million, $45 million, and $65 million, respectively. See the section titled “Management’s Discussion and Analysis of Financial Condition and Results of Operations—Non-GAAP Financial Measures” for information regarding our use of adjusted net (loss) income and a reconciliation of net loss to adjusted net (loss) income.
Industry Background
AI represents the next major evolution of technology, and its impact is potentially greater than the transformational shifts of the past. It is akin to a new industrial revolution delivering significant productivity gains, enabling new products that are poised to radically reshape and transform industries, and helping organizations become far more efficient.
Demand and Supply Side Factors Enabling AI
A combination of demand and supply side factors is enabling the massive and unprecedented growth of AI and powering the new industrial revolution.
Demand Side Factors
Supply Side Factors
Infrastructure is the Key Enabler of AI
We believe that specialized compute infrastructure is the key enabler that AI labs and AI enterprises need to train models, perform inference faster at scale, and accelerate their time to market. It has become generally accepted that there is a direct relationship between compute resources (FLOPs, training, and testing time) and model output quality. There has been a dramatic increase in the computational resources dedicated to training and hosting state-of-the-art AI models, with top models released by OpenAI, Google DeepMind, and Meta AI generally requiring exponentially greater quantities of compute compared to prior model generations to unlock performance gains. This principle applies for both training and inference: it states that the more compute power available, the better both the models and applications built on top of them become. To achieve greater compute capabilities and succeed in AI, AI labs and AI enterprises require specialized compute platforms that run at superior performance and efficiency and maximize the utilization of the underlying infrastructure.
Delivering Performant AI Infrastructure is Immensely Challenging
The highly specialized infrastructure that is required to unlock the potential of AI is immensely challenging to build and operate, especially at scale. The scale of the required infrastructure is staggering, with tens of thousands of GPUs, thousands of miles of high-speed networking cables, and hundreds of thousands of interconnects coming together to create “superclusters” for training and serving AI models, as well as hundreds of MW of power and petabytes of total storage. Depending on the configuration of the data center, a single 32,000-GPU cluster may require the deployment of approximately 600 miles of fiber cables and around 80,000 fiber connections. Acquiring the necessary high-performance components requires managing a complex global supply chain, and the data centers themselves need to be specifically designed for high-performance compute.
Software is Critical to Unlocking the Performance and Efficiency of AI Infrastructure
AI supercomputers are some of the most complex computing machines to have ever been created, requiring purpose-built software to unlock performance and efficiency at massive scale. These supercomputers require software to test and validate each component and to provision the infrastructure so that compute-intensive workloads can run with minimal downtime. Moreover, AI supercomputers often require multiple specialized orchestration frameworks to schedule complex workloads, which frequently need to be paired with substantial engineering resources. Once operational, these AI workloads put enormous pressure on infrastructure, frequently leading to high rates of component failures that cause training and inference jobs to fail, and thus require constant monitoring to manage and reduce failures and downtime.
AI Infrastructure Suffers from an MFU “Efficiency Gap”
Model FLOPS utilization is a measure of the observed throughput compared to the theoretical system maximum if it operates at peak FLOPs. The complexity of managing AI infrastructure means that a majority of the compute capacity embedded in GPUs is lost to system inefficiencies, with empirical evidence suggesting observed levels of performance within the 35% to 45% range. Minimizing the efficiency gap between the approximately 35% and the theoretical 100% represents a significant opportunity for unlocking AI infrastructure performance potential and therefore the improvement in quality of AI overall.
AI Requires a Purpose-Built Cloud
Generalized clouds operated by hyperscalers were not built to serve the specific requirements of AI. These clouds were created over a decade ago and were designed for general purpose use cases such as search, e-commerce, generalized web-hosting, and databases, and relied on CPU-based web-scale compute, and thus are not optimal for the high compute intensity requirements of AI.
Organizations need access to a fundamentally different GPU-based compute infrastructure that is built from the ground-up for AI and proven to perform at the scale and efficiency required to deliver AI. Cloud platforms therefore need to be fundamentally reimagined and purpose-built to deliver the capacity, performance, storage, and scalability requirements of AI workloads and increase the MFU of the underlying infrastructure.
Requirements of the AI Cloud
We believe the AI cloud requires a first-principles reimagining of how infrastructure is built, with the following requirements:
We believe the AI revolution requires a cloud that is performant, efficient, resilient, and purpose-built for AI. Today’s leading AI labs and AI enterprises demand high-performance and cost-efficient compute, storage, networking, and managed software services integrated into a cloud platform that enables them to bring their innovations to market faster by maximizing the utilization of their infrastructure and training and deploying models seamlessly and out-of-the box.
Our Solution
Our CoreWeave Cloud Platform is an integrated solution that is purpose-built for running AI workloads such as model training and inference at maximum performance and efficiency. It includes Infrastructure Services, Managed Software Services, and Application Software Services, all of which are augmented by our Mission Control and Observability software. Our proprietary software enables the provisioning of infrastructure, the orchestration of workloads, and the monitoring of our customers’ training and inference environments to enable high availability and minimize downtime. Built on a microservices-based architecture, the components of our platform are fully fungible and composable. Customers can configure their use of our CoreWeave Cloud Platform to best fit their needs. For instance, they can choose to bring their own storage or managed software services or run our respective solutions and choose the type and scale of deployment that best suits their workloads. This flexibility allows our customers to customize their use of our platform without compromising performance or efficiency. We have designed security as a fundamental component across our platform and technology stack. We leverage advanced security capabilities such as XDR and DLP, adhere to industry leading security standards such as SOC2 and ISO 27001, and employ our in-house information security teams to ensure that our customers operate in a secure environment.
The following is a summary of our Layered Architecture Stack.
CoreWeave’s Layered Architecture Stack
Infrastructure Services provide our customers with access to advanced GPU and CPU compute, highly performant networking (supported by DPUs), and storage.
Managed Software Services include CKS (a purpose-built-for-AI managed Kubernetes environment with a focus on efficiency, performance and ease of use), our flexible Virtual Private Cloud offering, and our Bare Metal service that runs Kubernetes directly on high-performance servers for maximum performance and efficiency.
Application Software Services build on top of our infrastructure and managed software services, integrating additional tools to further accelerate and improve training and inference for our customers. This includes SUNK, which allows customers to run Slurm-based workloads on top of Kubernetes and colocate jobs—including training and inference workloads—on a single cluster; CoreWeave Tensorizer, which significantly increases the efficiency of model checkpointing and enables high-speed model loading; and our inference optimization services.
Our purpose-built technology stack is augmented by our lifecycle management and monitoring software, Mission Control and Observability, and our advanced cluster validation, proactive health checking capabilities, and observability capabilities. Our AI cloud runs in a distributed network of 32 active purpose-built data centers, which are specifically engineered to support high intensity AI workloads with features including enhanced power, liquid cooling, and networking components, reinforcing the robustness of our entire technology stack. Our Third-Party Tooling and Solutions further enhance this flexibility by providing a composable architecture that allows customers to customize their solution by integrating additional third-party tools.
CoreWeave’s Data Center Network
Key Benefits to Customers
Through our CoreWeave Cloud Platform, we offer the following key benefits to customers:
CoreWeave Helps Bridge the MFU Gap
Competitive Strengths
Our key competitive strengths, which we believe set us apart from the generalized cloud providers in the industry, include:
in that our proprietary software and purpose-built infrastructure help to minimize the efficiency gap between the observed GPU cluster performance and the theoretical maximum GPU cluster performance, help to diversify their customer base, and help to get their technology in the hands of end customers faster. Our relationships with some of the world’s leading and most discerning AI labs and AI enterprises, including Microsoft and Meta, further position us as a cornerstone of the evolving AI ecosystem.
We serve some of the world’s leading AI labs and AI enterprises, which fosters a reinforcing cycle of continued innovation, as we are the first to experience and solve the challenges of the most compute intensive AI training and inference workloads. This data is a critical input into our engineering process and strengthens our technological differentiation as we continue to push the boundaries of what is possible in computing. This “economy of AI leadership” ensures we can continuously improve our CoreWeave Cloud Platform over time in a positive feedback loop, by enabling us to learn from any issues or misconfigurations detected across our large infrastructure base and apply those learnings to all our clusters, which in turn positions us as a leading-edge provider of choice for new customers.
Our Market Opportunity
The impact of AI will be immense for organizations of all sizes across all sectors, which will realize significant productivity gains and time and cost-efficiencies by deploying AI technologies. IDC estimates that AI will add nearly $20 trillion to global GDP by 2030.
We believe that CoreWeave will be one of the most sought-after platforms for the world’s AI workloads. We have estimated our total addressable market opportunity by evaluating the size of the large and rapidly growing AI compute software and infrastructure market that we are pioneering. According to Bloomberg Intelligence, the market for AI inference/fine-tuning, AI workload monitoring, and training infrastructure, including AI servers, AI storage, training compute, cloud workloads, and networking, will increase by over $300 billion from 2023 to 2028, growing at a CAGR of 38% from approximately $79 billion in 2023 to approximately $399 billion by 2028. This market opportunity is expected to include $330 billion related to training infrastructure, which includes AI servers, AI storage, training compute, LLM licensing revenue, cloud workloads, and networking; $49 billion related to inference infrastructure; and $20 billion related to workload monitoring, all of which are supportable by CoreWeave.
The AI compute market is centered on training the latest generations of models and increasingly leveraging those models to perform inference, including intensive inference workloads tied to complex reasoning tasks. In the long-term, as enterprise adoption of production workloads accelerates, we expect there to be an acceleration of inference workloads with varying levels of complexity and performance requirements. Our platform is ideally situated to capture all these workloads. Over time, as our geographically distributed compute base grows and includes older generations of infrastructure as current generations roll off existing contracts, we believe that the mix of hardware across generations that is optimized for different use cases in our compute base will ensure that we are optimally situated to capture all workload types, both inference and training.
Growth Strategies
Our principal growth strategies include: