Energy Vault BESS Infrastructure, 8 MW Crusoe Project, 2 GW Rack Scale Deal, and 12 Agreements (2025 to 2026)
Energy Vault’s AI Power Model, Bypassing Grid Bottlenecks
The core strategy driving data center infrastructure is shifting from securing land to securing power, forcing a convergence of energy and digital infrastructure development. Companies like Energy Vault are moving beyond selling standalone energy storage systems to delivering integrated “powered infrastructure” campuses. This model combines real estate, grid interconnection, and co-located battery energy storage systems (BESS) into a single, rapidly deployable package that directly addresses the multi-year delays associated with traditional utility grid upgrades, which have become the primary bottleneck for AI capacity expansion.
The Pre-2025 Model: Siloed Development Cycles
Prior to 2025, data center and energy project development operated on separate tracks, creating significant friction and long lead times. Data center operators would first secure a site and then enter lengthy and uncertain interconnection queues with local utilities, a process that could take several years. Energy developers, in parallel, built grid-scale storage assets primarily to sell ancillary services to the grid, without a direct, integrated offtake relationship with a specific industrial consumer like a data center.
The Post-2025 Shift: Integrated “Powered Shells”
The period from 2025 to 2026 marks the commercialization of the “powered shell” concept, driven by the intense AI power demand that cannot wait for conventional grid build-outs.
- Energy Vault’s groundbreaking on its Snyder, Texas AI campus in July 2026 for client Crusoe is the first major physical validation of this model. The project offers a complete powered infrastructure solution, targeting an aggressive Q 1 2027 commercial operation date for its initial 8 MW phase.
- This integrated approach allows AI companies to bypass utility queues by developing on sites with pre-existing or secured power access. The on-site BESS, such as Energy Vault’s B-VAULT™, manages grid integration, ensures power quality, and provides reliability for high-density compute workloads.
- The model is designed for rapid scalability. The Snyder campus starts at 8 MW but is designed for expansion to 25 MW and has a long-term site potential of 500 MW, offering clients a clear path for growth without repeating the entire site selection and grid interconnection process.
12 Energy Vault Agreements, Crusoe to Rack Scale (2025-2026)
Strategic partnerships have become the primary mechanism for executing the powered infrastructure model, creating a new value chain that aligns energy developers with data center operators. These alliances are structured to leverage specialized expertise, with energy firms managing power and grid complexity while data center partners focus on deploying compute capacity. This collaborative approach de-risks development and accelerates time-to-market for urgently needed AI infrastructure.
The Crusoe Partnership: A Commercial Proving Ground
The agreement between Energy Vault and Crusoe, announced in February 2026, serves as the commercial template for this new market segment. It established the framework for deploying Crusoe’s Spark™ modular data centers at Energy Vault-developed sites. The Snyder, Texas project is the first execution under this agreement, intended to prove the model’s speed and economic viability. For an 8 MW deployment, financial analyses suggest a potential annual EBITDA contribution between $12 million and $16 million for the data center operator once fully operational.
The Rack Scale Partnership: Scaling the Model to Gigawatts
Signaling a move from single projects to a programmatic build-out, Energy Vault signed a strategic partnership with Rack Scale Data Centers in late 2024. This agreement is far larger in scope than the initial Crusoe project, targeting the delivery of up to 2 GW of power to data centers. This partnership validates the market demand for the “powered shell” model at a gigawatt scale and positions Energy Vault as a key enabler of hyperscale AI deployments, moving well beyond its initial project-by-project approach.
Table: Energy Vault Strategic Partnerships and Project Announcements (2025-2026)
| Partner / Project | Time Frame | Details and Strategic Purpose | Source |
|---|---|---|---|
| Crusoe Cloud | Jul 2026 | Groundbreaking on an 8 MW powered AI infrastructure campus in Snyder, TX. Validates the “powered shell” model for rapid deployment of modular data centers. | Business Wire |
| Bay Wa r.e. AG | May 2026 | Acquisition of an 850 MW BESS portfolio in Japan. Establishes an operational platform in a key international market for potential replication of the AI infrastructure model. | Business Wire |
| Eskom | May 2026 | Development agreement for a 25 MW / 100 MWh Gravity Energy Storage System (GESS) in South Africa, continuing activity in its original core technology. | ess-news.com |
| Peak Energy | Feb 2026 | A 1.5 GWh supply agreement for sodium-ion batteries, aimed at securing the supply chain for future BESS deployments supporting AI data centers and reducing costs. | Business Wire |
| Rack Scale Data Centers | Dec 2024 | Strategic partnership to accelerate the delivery of up to 2 GW of power to data centers, scaling the powered infrastructure model to the hyperscale level. | Business Wire |
Texas ERCOT Market, Energy Vault’s 4 BESS and AI Projects
Texas has emerged as the primary geography for developing and testing the integrated energy and data center model. The state’s unique ERCOT market, characterized by high renewable energy penetration, frequent price volatility, and a favorable regulatory environment, creates the ideal conditions for co-locating battery storage with high-demand power users like AI data centers. This allows projects to capture value both from providing reliable power to the data center and from offering grid services to stabilize the broader ERCOT system.
Energy Vault’s Concentrated Texas Strategy
Energy Vault has established a significant and growing footprint in Texas, which serves as the foundation for its AI infrastructure strategy. Its portfolio is not just a collection of assets but a strategic network of power resources that can be leveraged for future data center developments.
- Snyder AI Campus: The 8 MW project with Crusoe is the first to directly monetize this strategy by co-locating a data center with a dedicated power solution.
- SOSA Energy Center: A 150 MW / 300 MWh BESS project in Madison County that began construction in January 2026, adding significant dispatchable capacity to its Texas portfolio.
- Mc Murtre BESS Project: The acquisition of this 175 MW / 350 MWh development project in March 2026 further deepens Energy Vault’s asset base in the state.
- Cross Trails BESS: A 57 MW project that reached FID in late 2024, demonstrating the company’s ability to bring BESS assets to operation in the ERCOT market.
8 MW to 2 GW, Energy Vault’s AI Infrastructure Scale-Up
The “powered infrastructure” business model has rapidly progressed from a conceptual solution to a commercially executed strategy, validated by project groundbreakings in 2026. While the underlying BESS technology is mature, the innovation lies in the commercial structure and integrated delivery. This model is now being proven at a project level, setting the stage for replication and scaling across a broader portfolio of energy development sites.
From Standalone Storage to Integrated Solutions
Between 2021 and 2024, Energy Vault’s primary focus was the development and deployment of energy storage hardware, including its novel gravity systems and more conventional BESS projects. These were largely treated as standalone grid assets. The strategic shift occurred in late 2024 and solidified through 2026, as the company pivoted to address the soaring infrastructure costs and power constraints of the AI industry. This transformed its business model from a hardware and software provider to an integrated infrastructure developer.
Commercial Validation at the Snyder Campus
The Snyder, Texas project is the critical validation point for this new model. Unlike previous announcements for standalone BESS projects, the Snyder campus represents the first physical construction of a fully integrated power and data center solution for an AI client. The aggressive timeline, with a targeted Q 1 2027 operation date, is designed to demonstrate that this model can deliver compute capacity significantly faster than the multi-year cycles typical of traditional hyperscale developments. Success at Snyder will serve as the commercial proof-of-concept for the much larger 2 GW pipeline with Rack Scale and other potential clients.
SWOT Analysis, Energy Vault First-Mover Advantage and Risks
Energy Vault’s strategic pivot has given it a first-mover advantage in the high-growth AI power infrastructure market, but this position comes with significant execution risks tied to the novelty of the integrated model. The company’s strengths are rooted in its ability to secure grid-connected sites and package them for AI clients, while its primary threat is the potential for larger, better-capitalized energy players or utilities to replicate the model at scale.
Table: SWOT Analysis for Energy Vault’s Powered Infrastructure Model
| SWOT Category | 2021 – 2024 | 2025 – 2026 | What Changed / Validated |
|---|---|---|---|
| Strengths | Focus on novel GESS technology and BESS integration software (Vault OS™). Building a project pipeline. | Pivoted to “powered infrastructure” model for AI. Secured major partnerships (Crusoe, Rack Scale). Building a large portfolio of BESS sites in key markets like Texas. | The company successfully translated its energy development expertise into a higher-value, integrated offering that solves a critical bottleneck for the AI industry, moving beyond hardware sales. |
| Weaknesses | Commercial viability of GESS technology was unproven at scale. High reliance on a few key projects for revenue. | Execution risk on the new, complex “powered shell” model. The Snyder project is the first of its kind for the company, and its success is not guaranteed. | The shift to a more complex, integrated model increases project management complexity and introduces new risks related to data center construction and client integration, moving beyond its core energy expertise. |
| Opportunities | Growing demand for grid-scale energy storage to support renewable integration. | Exponential growth in AI power demand creating a massive market for rapid, reliable power solutions. Multi-year utility interconnection queues create an opening for developer-led power. | The AI power bottleneck became the dominant market driver, creating a multi-billion-dollar opportunity that Energy Vault’s new business model is specifically designed to capture. |
| Threats | Competition from established BESS integrators. Fluctuating battery prices. Policy uncertainty around storage incentives. | Large utilities and IPPs (e.g., those with large land and grid portfolios like Xcel Energy or Duke Energy) could replicate the powered infrastructure model. A slowdown in AI demand could reduce the urgency for these projects. | The success of the model has made it a clear strategic target. The threat is no longer just from other storage providers but from any major energy or real estate player with grid-connected land assets. |
Energy Vault Q 1 2027 Test, Snyder Project’s Market Signal
The most critical factor for Energy Vault’s strategy over the next 18 months is its ability to execute the 8 MW Snyder, Texas project on schedule and on budget. Meeting the Q 1 2027 commercial operation target will serve as the definitive market signal, validating the “powered infrastructure” model and providing the commercial proof needed to unlock its larger pipeline of projects. Conversely, any significant delays would undermine confidence in this new approach.
If Snyder is On-Time and Successful
If the 8 MW project for Crusoe becomes operational in Q 1 2027, expect an immediate acceleration of Energy Vault’s AI infrastructure business. This would likely trigger a final investment decision (FID) on the 25 MW Phase 2 expansion at Snyder and advance tangible project development under the 2 GW Rack Scale partnership. A successful deployment will establish a powerful case study, attracting other AI and hyperscale clients seeking to avoid grid delays and cementing Energy Vault’s leadership in this emerging market segment.
If Snyder Faces Significant Delays
Should the project miss its Q 1 2027 target by a significant margin, it would raise questions about the viability and speed of the integrated model. Such a delay would suggest that the complexities of merging energy and data center construction are greater than anticipated. This would create an opportunity for competitors to catch up and could cause potential clients in the pipeline, such as those under the Rack Scale agreement, to hesitate or seek alternative solutions from more established infrastructure developers.
The questions your competitors are already asking
This report covers one angle of Energy Vault’s commercial trajectory. The questions that matter most depend on your work.
- Other companies building powered shells for data centers
- Data center power constraints in Virginia
- Sodium-ion battery supply for grid storage
- Crusoe modular data centers power source
This report does not answer these. Enki Brief Pro does.
Your question, your angle, your framework. SWOT, PESTL, scenario modelling. The same niche depth, built around the decision your work actually depends on.
