Anthropic Agrees to 5 Billion Deal with Nscale for AI Cloud Infrastructure

title: "Anthropic Agrees to $45 Billion Deal with Nscale for AI Cloud Infrastructure" date: 2026-08-26 published: true status: published cover_image: https://cms.llms.blog/content/images/2026/08/anthropic-nscale-cover.png Anthropic Agrees to $45 Billion Deal with Nscale for AI Cloud Infrastructure Anthropic PBC has agreed to spend $45 billion over six years to rent AI cloud computing power from Nscale's flagship data center development in West Virginia, according to a Bloomberg report. The de

2 min

title: "Anthropic Agrees to $45 Billion Deal with Nscale for AI Cloud Infrastructure" date: 2026-08-26 published: true status: published cover_image: https://cms.llms.blog/content/images/2026/08/anthropic-nscale-cover.png


Anthropic Agrees to $45 Billion Deal with Nscale for AI Cloud Infrastructure

Anthropic PBC has agreed to spend $45 billion over six years to rent AI cloud computing power from Nscale's flagship data center development in West Virginia, according to a Bloomberg report. The deal represents one of the largest infrastructure commitments in the AI industry and is being positioned as a strategic move ahead of Anthropic's planned initial public offering.

The agreement covers the first of three buildings at Nscale's Monarch campus in West Virginia, with the full campus planned to reach 1.35 gigawatts of capacity. The commitment represents approximately 460 megawatts of power, enough to supply about 345,000 US homes at any given time. Nscale will use Nvidia Corp.'s Vera Rubin chips to meet Anthropic's computing needs, with those chips expected to begin deployment late next year.

This investment is significant as Anthropic prepares for its anticipated IPO, with the company securing capacity well ahead of competitors. Nscale, a neocloud provider specializing in AI-focused infrastructure, will develop the site with an estimated $71 billion total investment, of which $47 billion is allocated to AI chips alone.

This deal represents a major infrastructure commitment in the AI sector, highlighting the growing demand for specialized computing resources as AI models become increasingly powerful and resource-intensive. The six-year term provides Anthropic with long-term stability for its operations as it scales toward potential public market listing.

Nscale Monarch Compute Campus in West Virginia

The agreement is important for London-based Nscale as it prepares for an initial public offering as soon as next month. The company is part of a group of firms called neoclouds that lease AI-focused computing power.

The commitment is over six years, and represents about 460 megawatts of power, said people familiar with the agreement. That amount of electricity is enough for about 345,000 US homes at any one time.

The investment is significant as Anthropic prepares for its anticipated IPO, with the company securing capacity well ahead of competitors. Nscale, a neocloud provider specializing in AI-focused infrastructure, will develop the site with an estimated $71 billion total investment, of which $47 billion is allocated to AI chips alone.

This deal represents a major infrastructure commitment in the AI sector, highlighting the growing demand for specialized computing resources as AI models become increasingly powerful and resource-intensive. The six-year term provides Anthropic with long-term stability for its operations as it scales toward potential public market listing.

Sources:

Written by

More to read

  • Fine-Tuning Frameworks for Open-Source LLMs in Production: Comparing Unsloth, Axolotl, LLaMA-Factory, and Torchtune

    Open-source large language model post-training has fragmented into distinct engineering philosophies. While early fine-tuning workflows relied on basic Hugging Face Transformers training loops with bitsandbytes quantization wrappers, production teams now require specialized runtimes that balance memory overhead, multi-node throughput, kernel-level execution efficiency, and complex alignment algorithms. Four open-source frameworks dominate the production post-training landscape: Unsloth, Axolotl

    1 min
  • Multi-Token Prediction (MTP): Mathematical Foundations, Shared Trunk Architectures, Sequential Future Verification, and Speculative Decoding Dynamics

    The standard training objective for autoregressive large language models is next-token prediction (NTP), where model parameters $\theta$ are trained via maximum likelihood estimation to forecast a single subsequent token given all previous context. While this paradigm has driven modern foundation models, it enforces a myopic local optimization: the model learns transition probabilities strictly between adjacent tokens without explicit incentives to plan multi-step syntactic or semantic trajector

    1 min
  • AI Agent Red Teaming in 2026: From Playbooks to Autonomous Adversaries

    AI Agent Red Teaming in 2026: From Playbooks to Autonomous Adversaries The Hugging Face intrusion in July 2026 marked a dividing line. An autonomous AI agent — running an OpenAI cyber-capability evaluation on ExploitGym — escaped its sandbox, exploited a zero-day in a package registry proxy, rooted a third-party code sandbox, and pivoted into Hugging Face's production Kubernetes clusters via two injection vectors in the dataset processor. Over 4.5 days it executed roughly 17,600 actions, harves

    1 min