CoreWeave Joins Nasdaq-100 Index After Rapid Growth

๐กCoreWeave's rapid rise to the Nasdaq-100 highlights the massive demand for specialized AI cloud infrastructure.
โก 30-Second TL;DR
What Changed
CoreWeave inclusion in Nasdaq-100 takes effect on June 22.
Why It Matters
CoreWeave's inclusion signals the continued dominance of AI infrastructure providers in the public markets, highlighting the massive capital flow into GPU-heavy cloud services.
What To Do Next
Evaluate CoreWeave's GPU availability and API pricing compared to hyperscalers for your next large-scale model training project.
Key Points
- โขCoreWeave inclusion in Nasdaq-100 takes effect on June 22.
- โขThe company pivoted from Atlantic Crypto mining to AI cloud services.
- โขRapid growth achieved within 15 months of the March 2025 IPO.
๐ง Deep Insight
Web-grounded analysis with 18 cited sources.
๐ Enhanced Key Takeaways
- โขCoreWeave was founded in 2017 as Atlantic Crypto in New Jersey by Michael Intrator, Brian Venturo, and Brannin McBee.
- โขThe company secured a significant $100 million investment from Nvidia in April 2023, followed by $2.3 billion in debt financing in August 2023, collateralized by Nvidia H100 GPUs.
- โขCoreWeave reported revenue of $6.23 billion and 130% growth over the last twelve months as of Q1 2026, with a market capitalization of $52.23 billion.
- โขIts infrastructure leverages NVIDIA BlueField-3 DPUs for offloading networking, management, storage, and security services, and utilizes NVIDIA Quantum-2 InfiniBand networking for high-performance GPU-to-GPU AI compute fabric.
- โขCoreWeave signed a multi-year agreement with Anthropic to support workloads for the Claude family of AI models.
๐ Competitor Analysisโธ Show
| Provider | H100 Price/hr (On-demand) | Billing Model | Key Features |
|---|---|---|---|
| CoreWeave | $4.76 (PCIe), ~$6.16 (HGX node normalized) | GPU, CPU, RAM, storage billed separately; discounts for multi-year commitments | Bare-metal, Kubernetes-native, NVIDIA Blackwell/Hopper/Ada GPUs, liquid cooling, InfiniBand interconnects, NVIDIA BlueField DPUs |
| Spheron | $1.33 (SXM) | Pay-as-you-go | Cost-effective, no commitment, multi-GPU support |
| RunPod | $1.99 - $2.39 | Pay-as-you-go | Developer-friendly, GPU renting, Kubernetes support |
| Lambda | $2.49 | On-demand or reserved (3yr for $1.84/hr) | Enterprise-grade, dedicated GPU, supports deep learning research |
| Paperspace | $5.95 | Managed ML workflows | End-to-end ML platform, notebooks, model repositories, deployment tools |
| Vast.ai | ~$1.87 (marketplace) | Pay-per-minute | Marketplace pricing, custom setup, multi-GPU support |
๐ ๏ธ Technical Deep Dive
- Architecture: Kubernetes-native architecture designed for large-scale, GPU-intensive tasks, providing bare-metal performance by eliminating virtualization overhead.
- GPUs: Offers a wide range of NVIDIA GPUs, including Blackwell, Hopper, and Ada Lovelace architectures, specifically H100, A100, RTX PRO 6000 Blackwell Server Edition, GB200 NVL72, HGX B200, RTX A5000, and A6000.
- Networking: Leverages NVIDIA BlueField-3 DPUs to offload and accelerate networking, management, storage, and security services, freeing CPU resources. Utilizes NVIDIA Quantum-2 InfiniBand networking for GPU-to-GPU AI compute fabric, offering low-latency and high-bandwidth data transfer.
- Storage: Provides CoreWeave AI Object Storage (S3-compatible), Dedicated VAST Storage, Distributed File Storage (POSIX shared filesystem), and Local Storage for various AI/ML data requirements.
- Cooling: Incorporates closed-loop, direct-to-chip liquid cooling in many data centers to manage heat generated by high-density GPU clusters, improving performance and reducing energy consumption.
- Services: Offers CoreWeave Kubernetes Service (CKS) for managed Kubernetes on bare metal, Slurm on Kubernetes (SUNK) for batch workloads, serverless and dedicated inference options, and CoreWeave Sandbox for ephemeral compute environments.
- Performance Claims: Claims up to 20% higher GPU cluster performance and up to 54% lower Total Cost of Ownership (TCO) compared to traditional hyperscalers for AI workloads.
๐ฎ Future ImplicationsAI analysis grounded in cited sources
โณ Timeline
๐ Sources (18)
Factual claims are grounded in the sources below. Forward-looking analysis is AI-generated interpretation.
Weekly AI Recap
Read this week's curated digest of top AI events โ
๐Related Updates
AI-curated news aggregator. All content rights belong to original publishers.
Original source: The Next Web (TNW) โ