- Specialization
- Distributed Computing
- Business
- Compute Hosting
- Market
- Software Development
- Ecosystem
- Neoclouds
- Customers
Enterprisesand 1 more customers valuesLess
- Developers
More details
- Organization type
- Commercial Company
- Ecosystem
- Neoclouds
- Business
- Compute Hosting
- Specialization
- Distributed Computing
- Market
- Software Development
- Customers
- Enterprises, Developers
- Delivery models
- Managed Compute
- Revenue models
- Usage Based
- Listing status
- Not classified
- Latest equity round
- Not classified
- Reported funding amount
- Not classified
About
CoreWeave builds an AI-native cloud platform combining GPU and CPU compute, storage, networking, managed services, and tools for model training, inference, agent development, and HPC. It sells to AI labs, startups, and global enterprises, including OpenAI, Mistral AI, and IBM, differentiating through vertically integrated, purpose-built infrastructure, high-speed interconnects, optimized runtimes, and expert support.
Market
CoreWeave competes in cloud infrastructure and GPU cloud services for AI training, inference, and high-performance computing. It positions itself as an AI-native, vertically integrated alternative to general-purpose hyperscalers, differentiating through purpose-built GPU infrastructure, managed Kubernetes, high-performance compute and storage, specialized operational support, and flexible public or dedicated deployments. Its principal competition includes the major hyperscalers as well as specialized GPU-cloud providers such as Lambda and Crusoe.
CoreWeave primarily serves AI model developers, hyperscalers, AI-native startups, and enterprises with demanding large-scale training and inference workloads. Relevant buyers include CTOs, CIOs, machine-learning and platform-engineering leaders, and public-sector or defense technology organizations that need specialized GPU capacity and operational support.
At a Glance
Problem
AI companies need enormous, specialized computing capacity to train, fine-tune, and serve increasingly complex models, but conventional infrastructure leaves them managing the complexity of acquiring scarce GPU capacity, scaling systems, and operating performance-sensitive workloads. CoreWeave positions this as a problem of making supercomputing accessible and managing the complexities of AI growth; its inference materials specifically identify operational overhead, changing traffic patterns, and workload economics as distinct challenges.
The main use case is business-critical generative-AI infrastructure, especially production inference: running models reliably and economically as demand fluctuates. This is a costly pain because customers must match expensive GPU capacity and system architecture to unpredictable workloads, while maintaining low latency and high utilization. CoreWeave’s customers include major AI and technology companies using the platform for these workloads.
Product / Service
CoreWeave provides an AI-native cloud and vertically integrated infrastructure platform purpose-built for AI. Its offering combines high-performance GPU compute, storage, a Kubernetes API, and managed Kubernetes for building, training, and deploying AI applications. The company also provides multiple inference paths designed for different operational requirements, traffic shapes, and economic profiles, allowing customers to choose between more managed convenience and more direct infrastructure control.
The delivery model is a specialized cloud service backed by the technology, tools, and teams needed to operate AI workloads at scale. By abstracting infrastructure complexity while supplying specialized compute, CoreWeave aims to let AI developers move faster, scale capacity more predictably, and improve the performance and economics of training and inference. Its vertically integrated stack is designed particularly for complex inference workloads.
Market
CoreWeave competes in cloud infrastructure and, more specifically, the specialized AI-cloud and GPU-cloud segment. Its competitive set includes large general-purpose cloud platforms such as AWS, Google Cloud, and Microsoft Azure, as well as AI infrastructure and orchestration providers such as Anyscale; Anyscale’s own positioning references execution across AWS, GCP, Azure, Nebius, and CoreWeave. CoreWeave differentiates through hyperspecialization in generative AI, managed Kubernetes, high-performance systems, and an integrated operating model rather than treating AI as one workload among many.
CoreWeave is well beyond pre-revenue: it completed its Nasdaq listing under ticker CRWV in March 2025, projected $4.9 billion to $5.1 billion of 2025 revenue, and had reported trailing-twelve-month revenue of $6.23 billion as of March 31, 2026. Its traction is reinforced by large customer commitments, including an expanded $21 billion agreement with Meta to scale inference workloads through 2032 and an OpenAI agreement with contract value of up to $11.9 billion, alongside relationships with customers such as IBM, Cohere, and Mistral AI.
Founders & Leadership
Funding History
Magnetar
Magnetar
Magnetar
Magnetar
Magnetar, Blackstone
Fidelity Management & Research Company, Morgan Asset Management, Nat Friedman, Daniel Gross, Goanna Capital, Zoom Ventures
Coatue
Blackstone, Magnetar
JPMorgan Chase, Goldman Sachs, Morgan Stanley
Stack Capital Group
OpenAI
Morgan Stanley, MUFG
Jane Street
Recent News
CoreWeave announced an India expansion in collaboration with AdaniConneX, planning 240 MW of data-center capacity across three buildings at the Taloja campus in Navi Mumbai. CoreWeave is to be the sole tenant, with an option to add another 240 MW; the first phase is expected to be available in mid-2028.
CoreWeave announced it will expand its compute portfolio with NVIDIA Vera CPUs, designed for AI-agent workloads. The company said Vera will run bare metal under the same CoreWeave platform and consumption models as its other compute offerings.
CoreWeave announced the availability of NVIDIA Vera Rubin NVL72 on its cloud, with Cognition as the first customer anywhere running production workloads on the system. Cognition reported up to 4.8 times greater total token throughput for its SWE-2 inference workloads compared with a GB200 NVL72 baseline.
CoreWeave launched Forge, a development layer for training, inference, evaluation and agent development in one connected environment. MasterClass and Canva were already building on Forge at the time of the announcement.
Ennoble Care selected CoreWeave Cloud for its clinical AI inference workloads, deploying NVIDIA RTX PRO 6000 Blackwell Server Edition capacity through CoreWeave Kubernetes Service. The arrangement provides dedicated, reserved GPU capacity for the home-based care provider.
Harell Data selected CoreWeave Cloud under a multi-year agreement for training, fine-tuning and inference on its secure platform for proprietary biotech datasets. CoreWeave said training runs execute on its NVIDIA GPUs while the raw data stays in place.
CoreWeave priced its private offering of $3.7 billion aggregate principal amount of 2.875% convertible senior notes due 2033, upsized from the previously announced $3.0 billion. The notes, which carry an initial conversion price of approximately $97.85 per share, are scheduled to settle on September 22, 2026.
CoreWeave established an at-the-market offering program allowing it to sell up to 35,000,000 shares of Class A common stock from time to time through a group of major banks as sales agents. The program provides ongoing financing flexibility and supports CoreWeave's objective of moving toward an investment-grade credit profile.
CoreWeave announced its intention to offer $3.0 billion aggregate principal amount of convertible senior notes due 2033 in a private offering, with an option for initial purchasers to buy up to an additional $500 million. The company said proceeds would be used partly to fund capped call transactions and for general corporate purposes supporting its AI infrastructure buildout.
CoreWeave announced the bring-up of a multi-rack NVIDIA Vera Rubin NVL72 cluster on CoreWeave Cloud, connecting hundreds of NVIDIA Rubin GPUs as a single scale-out cluster for agentic AI training and inference. It also introduced two new CoreWeave AI Object Storage capabilities: cross-region write acceleration and a no-fee Archive tier.
CoreWeave launched Physical AI Field Engineering, an offering that embeds CoreWeave engineers with customer teams to build, validate, and deploy AI across the full engineering lifecycle using customers' proprietary data. The service is built on the team and methods CoreWeave acquired with Monolith AI and has been applied across more than 100 engineering projects in automotive, aerospace, and robotics.
Parallel Works and CoreWeave announced the deployment of a fully managed AI and high-performance computing platform for DARPA's NODES program, pairing Parallel Works' ACTIVATE platform with CoreWeave's AI-native cloud. The environment includes a dedicated reservation on NVIDIA HGX H100 systems with enterprise-scale storage and NVIDIA Quantum InfiniBand networking to accelerate biological modeling and research.
CrowdStrike introduced SafeMind, a family of purpose-built security models and harnesses designed to operate natively in the Falcon platform. Developed with NVIDIA and supported by CoreWeave infrastructure, SafeMind combines offensive and defensive models in a continuous loop.
PagerDuty’s August 27 quarterly results release identified Anthropic, Banco Pichincha, CoreWeave, Delivery Hero, Kawasaki Heavy Industries, and Palo Alto Networks among its “lands and expands” customers. The release also listed PBC, but did not provide individual contract values or distinguish new customer wins from expansions.
CoreWeave announced that Rescale will expand its cloud ecosystem to CoreWeave Cloud for AI and simulation workloads supporting next-generation product development. The integration with CoreWeave Kubernetes Service will give Rescale customers access to CoreWeave’s AI-optimized cloud for generative AI and high-performance computing workloads.
CoreWeave announced a multi-year agreement with Hudson River Trading (HRT). HRT will use CoreWeave’s AI cloud platform, including NVIDIA Vera Rubin infrastructure and Spectrum-X networking, for AI-driven trading research and model development.
In its second-quarter update, CoreWeave identified Bentley Systems, Caterpillar, Grammarly, Isomorphic Labs, and Sunday Robotics as leading enterprise and AI customers. It also reported expanded relationships with Cognition, Databricks, Hudson River Trading, Periodic Labs, Rescale, and Runway ML.
CoreWeave announced new capabilities for running AI workloads across cloud environments, including CoreWeave Interconnect and SUNK Anywhere. It also launched agentic AI capabilities including CoreWeave ARIA and CoreWeave Sandboxes for experimentation, reinforcement learning, agent tool use, and model evaluation.
CoreWeave announced the closing of a $2.6 billion delayed-draw term loan facility to finance HPC-backed infrastructure dedicated to customer contracts. The approximately five-year facility was meaningfully oversubscribed and was arranged by JPMorgan and Mitsubishi UFJ Financial Group.
IMC announced that it will significantly scale its use of CoreWeave’s AI cloud platform, including CoreWeave Kubernetes Service, to support next-generation research infrastructure. IMC has operated CoreWeave clusters in production since 2025 and is expanding to a substantially larger compute footprint for quantitative research.
CoreWeave reported first-quarter 2026 results and a $99.4 billion revenue backlog as of March 31, 2026. The company also launched CoreWeave ARENA for production-ready workload evaluation.
Jane Street committed approximately $6 billion to use CoreWeave’s AI cloud platform and invested $1 billion in CoreWeave equity, expanding the companies’ relationship.
CoreWeave announced the closing of an $8.5 billion financing facility to support its expansion of AI cloud infrastructure.
CoreWeave expanded its partnership with OpenAI by up to $6.5 billion, bringing the total value of the companies’ agreements to approximately $22.4 billion for AI training and deployment.
Active Roles
306Business Model
CoreWeave monetizes access to AI cloud infrastructure and related services, with GPU and CPU capacity priced by usage, including on-demand and spot hourly pricing, alongside storage and other platform offerings. A substantial portion of revenue comes from committed or reserved-capacity customer contracts; its 2025 filing said customer commitments represented 98% of recognized revenue.
Products
Customers
Tech Stack
Similar Companies
Competitors
Key Investors
NVIDIA, Morgan Stanley, BlackRock, Coatue, Fidelity Investments, Jane Street, JPMorgan Chase, Blackstone