CoreWeave's Multibillion Deal with Anthropic

CoreWeave secures $B+ deals with Anthropic/Meta: AI cloud capacity surge imminent
30-Second TL;DR
What Changed
Multibillion-dollar contract signed with Anthropic
Why It Matters
These massive deals position CoreWeave as a key AI infrastructure provider, potentially accelerating model training for top labs amid GPU shortages. It signals growing demand for specialized cloud compute in AI.
What To Do Next
Contact CoreWeave sales to explore GPU cluster pricing for your AI workloads.
Key Points
- •Multibillion-dollar contract signed with Anthropic
- •$21 billion deal with Meta for computing power
- •CEO Michael Intrator confirmed deals publicly
Deep Insight
AI-generated analysis for this event — not the original article.
Enhanced Key Takeaways
- •The deal with Meta represents a significant expansion of CoreWeave's infrastructure footprint, specifically designed to support the training and inference requirements of Llama 4 and subsequent model iterations.
- •CoreWeave is leveraging its specialized GPU cloud architecture to provide Anthropic with low-latency access to H200 and Blackwell-based clusters, critical for scaling the Claude model family.
- •These agreements solidify CoreWeave's position as a primary 'hyperscaler-alternative,' shifting the company's business model from general-purpose cloud services to dedicated, long-term capacity reservations for top-tier AI labs.
Competitor Analysis
- CoreWeave
- GPU-specialized, bare-metal performance
- AWS (EC2 UltraClusters)
- General-purpose cloud + AI
- Microsoft Azure (AI Infrastructure)
- Enterprise integration + AI
- CoreWeave
- Long-term capacity reservations
- AWS (EC2 UltraClusters)
- On-demand/Reserved/Spot
- Microsoft Azure (AI Infrastructure)
- Enterprise agreements/Reserved
- CoreWeave
- NVIDIA H100/H200/Blackwell
- AWS (EC2 UltraClusters)
- NVIDIA H100/Trainium/Inferentia
- Microsoft Azure (AI Infrastructure)
- NVIDIA H100/Maia (custom)
| Feature | CoreWeave | AWS (EC2 UltraClusters) | Microsoft Azure (AI Infrastructure) |
|---|---|---|---|
| Primary Focus | GPU-specialized, bare-metal performance | General-purpose cloud + AI | Enterprise integration + AI |
| Pricing Model | Long-term capacity reservations | On-demand/Reserved/Spot | Enterprise agreements/Reserved |
| Hardware | NVIDIA H100/H200/Blackwell | NVIDIA H100/Trainium/Inferentia | NVIDIA H100/Maia (custom) |
Technical Deep Dive
- Infrastructure utilizes NVIDIA Blackwell B200 GPUs interconnected via high-bandwidth InfiniBand fabrics to minimize communication overhead during large-scale model training.
- Implementation relies on a proprietary software-defined networking (SDN) layer that allows for dynamic reconfiguration of GPU clusters based on specific workload requirements.
- The architecture supports multi-tenant isolation at the hardware level, ensuring data security for Anthropic and Meta while maintaining bare-metal performance characteristics.
Future ImplicationsAI analysis grounded in cited sources
Timeline
- 2023-04CoreWeave secures $221 million in Series B funding to expand GPU cloud infrastructure.
- 2023-08Company raises $2.3 billion in debt financing led by Magnetar Capital and Blackstone.
- 2024-05CoreWeave announces $1.1 billion Series C funding round at a $19 billion valuation.
- 2025-02CoreWeave secures $650 million in secondary share sales, further increasing valuation.
- 2026-04CoreWeave confirms multibillion-dollar deals with Anthropic and Meta.
Weekly AI Recap
Read this week's curated digest of top AI events →
AI-curated news aggregator. All content rights belong to original publishers.
Original source: Bloomberg Technology ↗
This is a summary, not the original. Read the source, or get the weekly briefing.
The weekly digest
One email a week. Unsubscribe anytime.