Crusoe Energy has raised over $3 billion at a $30 billion valuation, tripling its worth in less than a year. The funding round, co-led by Atreides Management and Valor Equity Partners with participation from Mubadala Capital, reflects explosive investor appetite for companies building large-scale AI data center infrastructure. The valuation jump from $10 billion in October 2025 to $30 billion in September 2026 underscores how critical physical computing capacity has become as artificial intelligence models grow larger and more power-hungry.

Why Is a Data Center Company Raising This Much Money?

The short answer: AI infrastructure requires enormous upfront capital, and long-term customer commitments make that investment bankable. Crusoe announced a five-year cloud computing contract with Jane Street Group valued at approximately $13 billion, making the quantitative trading firm its highest-profile customer to date. This contract was instrumental in attracting the latest round of funding because it provides revenue visibility that investors crave. Rather than relying on month-to-month GPU rentals, Crusoe now has a predictable, multi-year revenue stream that justifies the massive capital expenditure required to build data centers at gigawatt scale.

Jane Street's decision to commit $13 billion over five years signals a broader shift in how AI infrastructure is purchased. Quantitative trading firms like Jane Street process enormous amounts of data and require cutting-edge computing power to maintain their competitive edge. The company's willingness to lock in capacity years in advance demonstrates that demand for advanced AI infrastructure is spreading far beyond AI laboratories and cloud providers into financial services and other computationally intensive industries.

What Does Crusoe Actually Build?

Crusoe positions itself as a vertically integrated AI infrastructure company rather than a traditional cloud provider. The company combines power procurement, data center design and construction, GPU infrastructure, and cloud services within a single platform. This integrated approach aims to reduce delays between securing electricity and making AI compute available to customers. Traditional data center development often requires separate partnerships with utilities, real estate firms, construction companies, and cloud providers, which can slow projects significantly. Crusoe's model attempts to compress these timelines by controlling more of the supply chain internally.

The company's flagship project is its Abilene, Texas campus, which has become a showcase for its development capabilities. The first phase, which became operational in 2025, was built on Oracle Cloud Infrastructure and includes Nvidia GB200 systems for AI training and inference workloads. Construction began in June 2024, and Crusoe energized the first two buildings within roughly one year, demonstrating the rapid development timelines that AI customers now demand. The completed eight-building plan is designed to support hundreds of thousands of GPUs on an integrated network fabric.

In March 2026, Crusoe announced an additional 900 megawatt AI factory campus in Abilene to support Microsoft's AI infrastructure needs. This project includes two new buildings and an onsite power plant, with the first building expected to be energized in mid-2027. Combined with Crusoe's existing infrastructure, the Microsoft development is expected to increase the broader Abilene site to approximately 2.1 gigawatts of total capacity. This illustrates how single customer relationships can now involve massive infrastructure commitments.

How to Understand Crusoe's Growth Trajectory

• Contracted Capacity: As of June 2026, Crusoe had reached 4.9 gigawatts of contracted AI infrastructure capacity across its data center projects and Crusoe Cloud platform, representing infrastructure already committed to customers rather than speculative future developments.
• Development Pipeline: The company's wider development pipeline exceeds 40 gigawatts, including contracted projects, locations under active tenant negotiation, and sites in advanced development stages, explaining why Crusoe requires continuous large capital infusions.
• Investor Participation: The latest funding round included co-leads Atreides Management and Valor Equity Partners, plus Mubadala Capital, the alternative-asset investment arm of Abu Dhabi's sovereign wealth fund, signaling international confidence in the AI infrastructure market.

The $30 billion valuation reflects investor confidence that Crusoe can execute on its ambitious expansion plans. The company raised $1.375 billion in its Series E round in October 2025, which valued it above $10 billion. The new round, completed in September 2026, represents a tripling of that valuation in less than a year. This acceleration is unusual even in the fast-moving AI sector and suggests that investors view long-term AI infrastructure contracts as sufficiently valuable to justify massive valuations.

The funding also highlights a fundamental shift in how AI infrastructure is financed and purchased. Rather than building data centers speculatively and hoping to attract customers, companies like Crusoe now secure multi-year customer commitments first, then raise capital to build the infrastructure. This model reduces risk for investors and provides clearer paths to profitability. Jane Street's $13 billion commitment essentially pre-funds a significant portion of Crusoe's expansion, making the company's growth trajectory more predictable than traditional infrastructure plays.

As AI models continue to grow in size and complexity, the demand for specialized data center capacity will likely intensify. Crusoe's ability to secure long-term contracts, execute rapid construction timelines, and integrate power procurement with infrastructure development positions it as a key player in the emerging AI infrastructure market. The company's $30 billion valuation, while striking, reflects the enormous capital requirements and revenue potential of building the physical foundation for artificial intelligence at scale.