Good morning, and welcome to Hut 8's third quarter 2026 financial results conference call. Joining us today are our CEO, Asher Genoot, and our CFO, Sean Glennan. Following the presentation, we will open the line for questions. This event is being recorded, and a transcript will be made available on our website. In addition to the press release issued earlier today, a full quarterly report on Form 10-Q is available at hut8.com, on our EDGAR profile at sec.gov, and on our SEDAR+ profile at sedarplus.ca. Unless otherwise indicated, all figures discussed today are in USD. Certain statements made during this call may constitute forward-looking statements within the meaning of applicable securities laws. These statements reflect current expectations and are subject to risks and uncertainties that could cause actual results to differ materially.
Certain key risks are detailed in our Form 10-K for the year ended December 31, 2025, and our other continuous disclosure documents. Except as required by law, we assume no obligation to update or revise any forward-looking statements. During the call, management may reference non-GAAP measures such as Adjusted EBITDA. We believe these metrics, alongside GAAP results, provide valuable insight into our performance. Reconciliations of GAAP and non-GAAP results are included in the tables accompanying today's press release, available on our website. I'll now turn the call over to Asher.
Thank you, Mark, and good morning, everyone. Over the past year, we have contracted 949 MW of IT capacity under long-term AI data center leases with or backed by investment-grade companies, representing approximately $27 billion of contracted revenue. Realizing that revenue and the returns that we underwrote against it requires three disciplines. First, the quality of the power and the contracts that we originate. Second, our execution in delivering what we have contracted. Third, our rigor in financing. Those are the three principles we outlined for investors in May. In my remarks this morning, we'll address each in turn, origination, delivery, and financing. Sean will then review the results in detail. Before I turn to the quarter, I want to step back to the premise behind all three measures, because it explains why we sequence them in that order.
AI is delivered as software, but it is produced by physical infrastructure, power, land, transmission, buildings, cooling, equipment, and the people who operate them. The next generation of compute will be constrained not only by chips, but also by that infrastructure. Of everything that infrastructure requires, power is a scarcest input. That is why we are a power-first company. The input that is hardest to secure is the one we secure first. We developed the discipline of building within that constraint in Bitcoin mining, which required us to operate power-dense compute to manage cooling and uptime, and to procure and build quickly. AI is the largest opportunity we have yet encountered to apply that same discipline. Securing power, however, only earns the right to build. Converting it into contracted revenue depends on how efficiently we build, and that is the second part of what defines Hut 8.
I am convinced that the infrastructure behind AI can be built better than it is being built today, faster, at a lower cost, and without compromising on quality. Many of today's data centers resemble NASA rocket programs, remarkable engineering, but complex, expensive, and slow to deliver. Frankly, this is the mold that River Bend and Beacon Point Phase 1 follow. This approach makes sense when the build-out is small, but it will not work at the multi-gigawatt scale at which I expect Hut 8 to build. Our future builds, therefore, start with a different question, what each system must do and how it should be designed and delivered. We are rethinking builds from there with the aim of improving speed and cost efficiency, with quality held as a constant, not the variable. We have done this before.
Vega, a Bitcoin mining facility we designed in-house with direct liquid-to-chip cooling, was built in under six months. We believe we can apply the same approach to our future builds, and we believe that our ability to rethink builds to be more cost-efficient and shorten delivery times while maintaining quality through innovation will differentiate Hut 8 with customers. With that as context, I will turn to the quarter, beginning with origination, which is where the power is secured. For investors, origination is a measure of whether we control sufficient power on terms that support attractive commercialization opportunities to continue converting our pipeline into leases comparable to those at River Bend and Beacon Point. We originate through three channels, and we advanced each throughout the quarter. The first is front-of-the-meter greenfield development. Our pipeline now includes more than 6.4 GW under diligence and under exclusivity.
Capacity under exclusivity, the stage at which we have secured a defined path to ownership, increased by approximately 1.3 GW during the quarter as existing projects advanced and new projects were originated. Pipeline alone does not answer the question investors ask most often about that pipeline, which is how geographically diversified is it? This quarter, we are providing a breakdown of our front-of-the-meter greenfield pipeline by geography. The largest shares are in MISO, PJM, and SERC. Less than 5% is in ERCOT. That distribution reduces our regulatory and price exposure in any single market. It does not, however, diminish the value of our existing Texas interconnects, which we believe have become more valuable as ERCOT's revised interconnection process has lengthened the time required to secure approvals for new interconnections.
The second channel is behind-the-meter development, which allows us to unlock power and accelerate delivery where traditional grid timelines are constrained, and which builds on years of experience developing and operating such projects. We are actively evaluating opportunities across our campuses. Consistent with our disciplined approach, however, we do not include that capacity in our reported development pipeline until a contract is executed. River Bend illustrates that point. Beyond the current phase under construction, we continue to develop the site. We have cleared 184 acres for campus expansion and are advancing site preparation now, so that construction can begin promptly once the next phase is commercialized and approvals are secured. The third channel is corporate development, in which we acquire powered assets.
Our approach here is the one that produced Beacon Point, where we secured the power, underwrote the site on Bitcoin mining as its initial use, and applied our greenfield development capabilities to advance the site and convert it into an AI data center campus. This quarter, corporate development yielded another opportunity as we entered into a definitive agreement to acquire from affiliates of Quoin two energized data center sites in West Texas, with up to 230 MW of utility capacity upon full energization, subject to final power approvals and infrastructure development. Second, delivery. Each lease we sign obligates us to deliver capacity by a contracted date, and the lease revenue begins only on delivery. For investors, delivery is therefore the measure of whether the capacity we have contracted begins generating rent on the dates to which we committed, and it is our central operating priority.
Construction at River Bend and Beacon Point Phase 1 remains on track to meet our delivery targets, and both projects are funded entirely at the project level. At River Bend, more than 1,200 people are on site. Structural steel, roof decking, and building foundations are complete on all four data halls. The slab at Data Hall 1 is also complete, and the remaining three data halls will follow through year-end. Roofing and precast wall panels have also been installed at Data Hall 1, with the remaining halls expected to follow through by mid-December, reducing the exposure of interior electrical and mechanical work to weather-related disruption.
We expect to energize the substation in phases, and equipment installation has begun in Data Hall 1, starting with Vertiv IT power skids and SmartRun overhead modulars. At Beacon Point Phase 1, ground improvements are approximately 95% complete. Building foundation work and underground utilities installation are underway.
The steel erection is scheduled to begin this month. We have also begun foundation work for the substation. For Beacon Point Phase 2, civil design and surveying are underway. We also amended the Beacon Point Phase 2 lease with our customer, extending its term from 15 to 17 years. The extension serves both parties. For our customer, it secures capacity over a longer horizon, which reflects the durability of demand for power. For Hut 8, it adds two years of contracted revenue for lenders to underwrite, which should help offset potentially higher financing costs as we look to finance Phase 2. The amendment applies to Phase 2 only. Phase 1 remains on its existing terms, and the construction and financing framework for the campus is unchanged.
Delivery also depends on the support of the communities in which we build, which we treat as a precondition of development rather than as an obstacle to be managed. River Bend demonstrates what is required to convert power-first approach into a real project. Power, transmission, land, construction, partners, and community alignment must come together in the right order. That is what more than 300 investors, analysts, media, government stakeholders, and community members saw when they visited River Bend last month. I joined Governor Jeff Landry of Louisiana and leaders from Anthropic, Entergy Louisiana, Broadcom, Vertiv, and Jacobs to discuss the future of American AI infrastructure, and those discussions reached one conclusion: demand, supply, delivery, and policy are inseparable. Sound technical plans matter, but durable projects are built transparently in partnerships with the people who live and work nearby. In Louisiana, that partnership has taken specific form.
We are investing approximately $16 million in a new water well and nearly 8 miles of water infrastructure, none of which will be used to cool the data center, and which will be transferred to the parish at no cost to local taxpayers and can serve more than 4,000 households and 200 local employers. We also made a $10 million payment in lieu of taxes, bringing tax revenue to West Feliciana Parish one year earlier than it otherwise would have arrived, supporting local government, schools, and law enforcement, including $7,000 stipends for school employees in each of the next two school years and payments of up to $8,000 for eligible sheriff office personnel. At Beacon Point in Texas, the commitments differ in form, but not in principle.
We are funding the electrical upgrades that our campus requires so that those costs are not shifted to residential ratepayers, and the campus uses closed-loop cooling design to minimize local water demand. Across both phases, we anticipate approximately 1,900 construction jobs and more than 200 permanent positions, supported by partnerships with local educators and workforce organizations. We make these commitments because we believe a successful project must create lasting value, not only for our customers and shareholders, but also for the communities that host it. We uphold this philosophy by investing in the infrastructure, workforce, and partnerships that allow our projects and the communities around them to grow together. Third, financing. For investors, how we finance our projects is a measure of our ability to limit risk to the parent balance sheet and avoid dilution for existing shareholders.
Two of the main questions we've been asked this quarter get directly to that point. How are our current builds progressing, and might we need to issue additional equity to support them? The answers are straightforward. We remain on schedule, and we do not expect to issue additional equity to support River Bend or Beacon Point. River Bend and Beacon Point Phase 1 are funded through completion with investment-grade project bonds of $3.25 billion for River Bend and $4.25 billion for Beacon Point Phase 1, each non-recourse to Hut 8. We were able to raise investment-grade financing for both projects through a comprehensive approach to risk mitigation. Operating in Bitcoin mining taught us how quickly markets can change, and inflation and interest rate expansion are present realities. We cannot control the market, but we can control who we secure in advance and how we contract.
We work from first principles to understand what every component must do and what it should cost. At River Bend and Beacon Point Phase 1, all long-lead equipment has been ordered, and all major contracts have been signed, locking in the largest components of cost ahead of further market movement. Risk mitigation extends beyond procurement. Our principal equipment and construction partners, Vertiv and Jacobs, are investment-grade counterparties. Comprehensive insurance coverage is in place at both sites, providing an additional layer of protection against covered losses that could affect the projects and, in turn, bondholders. And the triple net lease structure makes tenants responsible for operating expenses, taxes, and insurance, which further supports this framework. A measure of the strength of our financings is that both bonds have traded well in the secondary market.
That matters because capital has become scarcer across today's markets, and we believe scarcity produces a flight to quality. Investor interest in future issuance remains strong. In a tighter market, we believe contract quality and the flexibility to access multiple debt capital markets gives us a competitive advantage. As we evaluate financing for future projects, we will start from first principles with a clean slate, seeking the lowest cost of capital and the highest free cash flow yield on a risk-adjusted basis. How we originate, how we deliver, and how we finance also answers the most direct question investors ask: What will drive the stock? We believe execution is how we create long-term shareholder value. We do not manage to the daily share price.
We manage to the delivery of River Bend and Beacon Point and to the conversion of our pipeline into contracted projects, because those are the outcomes within our control. The same long-term focus governs how we allocate capital. Every decision is evaluated against a single criteria, whether it creates long-term shareholder value. Alongside investing in growth, that criteria allows us to opportunistically create shareholder value by repurchasing our shares when we believe the stock trades at a discount to intrinsic value. We have acted on that conviction since last quarter by repurchasing approximately 3.3 million shares for approximately $250 million, funded with about 3,100 Bitcoin from our balance sheet. The repurchase met our criterion because we believe our shares do not fully reflect the value of the revenue we have contracted and the power capacity that we control.
Funding it with Bitcoin rather than with cash preserved liquidity and continued the reduction in parent-level Bitcoin exposure that we described in February. Underneath those measures is a simple proposition. We are builders, and we believe the AI era will reward the companies that bring power and compute online faster, more cost-efficiently, and without compromising on quality. That is the company we are building. The revenue we have contracted will be realized the way it was secured, by starting with power, underwriting carefully, preserving flexibility, and delivering. Those principles govern the three measures I have described, and those measures are how we expect to be judged in the quarters ahead. We believe they are the foundation of an enduring generational business at the intersection of energy and technology. With that, I will turn the call over to Sean.
Thanks, Asher. The financial story this quarter has four main components. First, the operating platform remained resilient. Revenue was $71.1 million for the third quarter, compared with $83.5 million in the prior year period. While Adjusted EBITDA increased to $33 million from $30.5 million. That result reflects the contribution from additional operating capacity and a disciplined approach to cost. Despite the sale of the Far North portfolio in February, a decline in average revenue per Bitcoin mined, and the planned runoff of businesses that are no longer central to our strategy. Second, the GAAP result was heavily affected by the fair value remeasurement of digital assets. We reported net income of $189.5 million, of which $131.1 million was attributable to Hut 8. That result included a $413.8 million gain on digital assets, reflecting the increase in the price of Bitcoin during the quarter.
To help investors assess operating performance alongside these price movements, we supplement our GAAP results with Adjusted EBITDA, a non-GAAP measure, and our segment results. Adjusted EBITDA should be considered together with, and not a substitute for, GAAP net income. Third, we continue to maintain cost discipline even as we scale the company. While we are making major investments to scale them to the company we want to become, we are doing so judiciously. The rigor around spend decisions has not changed since I started two years ago, when the company was in more of a startup phase. Compared with the prior year period, G&A, excluding stock-based compensation, per megawatt of development pipeline has declined, even as our development pipeline has nearly doubled. Last, but arguably most important, the balance sheet and capital position of the company remain strong. We have fully funded two construction stage projects.
We have ample liquidity to meet our obligations until our data centers begin generating cash flow, and that liquidity is now further supported by the $1.07 billion revolving credit facility that we closed in September, which provides low-cost, flexible, non-dilutive capital. We have confidence in the availability of capital for strong projects underpinned by robust contracts and thoughtful risk mitigation. Let me walk through the operating business. Compute. Compute revenue was $68.4 million in the third quarter, compared with $70 million in the prior year period. Headline revenue was broadly stable because the decline in average revenue per Bitcoin mined was largely offset by an increase in Bitcoin mined, reflecting additional operating capacity at Vega and the re-energization of Drumheller. That result demonstrates the enterprise's ability to scale revenue across a range of external environments.
Compute cost of revenue increased to $28 million from $22 million, primarily because we were operating more capacity and incurring the associated electricity costs. The increase tracks the increase in production, not a deterioration in operating efficiency. The important point is that we are producing more Bitcoin from a larger operating footprint, again, demonstrating the ability to scale in a cost-conscious manner. AI cloud revenue also increased during the quarter as new customer contracts came online. That growth was partially offset by the sale of the traditional cloud business in August. Power. Power revenue was $1.3 million in the third quarter, compared with $8.4 million in the prior year period. The decline was primarily the result of the sale of the Far North portfolio in February.
It is therefore a portfolio composition effect, not a deterioration in the remaining power business, which generates revenue primarily from managed services contracts. Digital infrastructure. Digital infrastructure revenue was $1.4 million in the third quarter, compared with $5.1 million in the prior year period. The decline primarily reflects the termination of the Vega co-location agreement after American Bitcoin exercised its purchase option for the miners at that site, together with the customer churn in the legacy CPU infrastructure business. The current digital infrastructure results do not yet reflect the contracted lease revenue from River Bend and Beacon Point. That revenue begins when the relevant capacity is delivered. The financial model is therefore moving through an intentional transition. The legacy revenue base is no longer the main growth driver, while the larger contracted infrastructure revenues remain tied to delivery milestones.
Adjusted EBITDA was $33 million, compared with $30.5 million in the prior year period. We use this measure to evaluate operating performance after excluding the effect of fair value remeasurement of digital assets and other one-time or non-cash items. As always, Adjusted EBITDA should be considered alongside GAAP results, not in place of them. Reported general and administrative expense was $99.6 million, compared with $25.9 million in the prior year period. Approximately $64.7 million of this quarter's G&A was stock-based compensation, including the effect of one-time long-term incentive awards. The quarter also included professional fees and approximately $6.2 million of incremental salaries and benefits as we added people to support energy origination, corporate development, and other growth functions. Excluding non-cash costs, the cash G&A was $34.8 million, which compares with $24.8 million last quarter and $19.5 million from a year ago.
We do not view all of that spending as recurring corporate overhead. A meaningful portion reflects the cost of building the organization required to originate, finance, construct, and operate multiple large campuses in parallel. At the same time, the cash component is real, and we are holding those investments to a clear standard. They must support one or more of the following outcomes over time: more high-quality power origination, faster project conversion, a lower cost of capital, better execution, or stronger operating leverage. The balance sheet requires an important distinction between corporate liquidity and project-restricted capital. At September 30, 2026, we had approximately $156.9 million of cash, $3.6 billion of restricted cash and cash equivalents, and $2.8 billion of restricted short-term investments.
The restricted balances are principally proceeds from the River Bend and Beacon Point financings, held in project accounts for construction, debt service reserves, and other specified project purposes. They are not excess corporate cash. Hut 8 also has about 1,361 unencumbered Bitcoin on the balance sheet as of November 1st. The company had approximately $7.6 billion of consolidated loans, notes payable, and other financial liabilities. Approximately $7.5 billion of that amount relates to the River Bend and Beacon Point project financings. Those obligations sit at project subsidiaries, are secured by the applicable project assets and accounts, and are non-recourse to Hut 8's parent company. Each project is designed to service its own debt from its own contracted lease cash flows. The quarter also included $105.5 million of interest expense and $61.8 million of interest income.
The interest expense reflects the construction period carry on the River Bend and Beacon Point financings. The interest income was generated by investing unused project proceeds in restricted short-term investments to partially offset that carry. We accepted that construction period interest cost because the projects are fully funded upfront. The parent-level financing architecture is separate and quite simple. In September, we added a $1.07 billion senior secured revolving credit facility to support corporate liquidity, development needs, and letters of credit. No borrowings were outstanding under that facility as of September 30, 2026. The facility gives us committed capacity without requiring us to draw debt before we need it. We also simplified the parent capital structure during the year. The Coatue Note converted into equity. The Coinbase facility was refinanced with the Bitcoin-collateralized FalconX term loan at a lower coupon, and the remaining TZRC Secured Promissory Note was repaid in October.
We are also planning to retire the FalconX term loan ahead of its maturity in April 2027. The objective is straightforward: maintain parent-level flexibility, match financing to the asset that generates the cash flow, and avoid using corporate equity where project-level capital is available on better terms. During the quarter, we sold approximately 954 Bitcoin for $74.8 million and used those proceeds to repurchase 953,802 shares for $74.7 million. Subsequent to quarter end, we repurchased additional shares under the $250 million program. In total, we repurchased approximately 3.3 million shares for approximately $250 million, funded with approximately 3,100 Bitcoin. Those repurchases converted a portion of the strategic Bitcoin reserve into equity at a price we believe did not reflect the contracted revenue and power capacity controlled by the company. They also continued the reduction of parent-level Bitcoin exposure that we had described previously.
Our board has also authorized an increase in our share repurchase program from $250 million to $1 billion. We view this as a long-term program that gives us the flexibility to act when opportunities arise over the coming years. We will be patient and judicious in how we deploy it, so that we maintain liquidity to run our business and pursue opportunistic growth. Stepping back, investors should evaluate Hut 8 through three layers. The first is the solid footing of the current operating platform, where the quarter demonstrated increased Bitcoin production, a growing AI cloud contribution, and Adjusted EBITDA above the prior year period despite lower revenue. The second is the contracted infrastructure platform, where River Bend and Beacon Point are being funded and delivered through project-level structures tied to long-duration customer commitments.
The third is the development platform, where we are advancing power opportunities while preserving flexibility on the application, the customer, and the financing structure. Our financial objective is not to maximize headline growth at any cost. It is to convert scarce power into financeable, contracted infrastructure while protecting parent-level liquidity and limiting unnecessary dilution. If we execute against that standard, the financial profile should evolve from a business with Bitcoin-driven volatility and construction period carry toward one with a larger base of contracted infrastructure cash flows.
Before we transition to Q&A, let me address what is likely to be the first question: the outcome of the governor's race in Texas and its implications for Beacon Point. We respect the election outcome and support the Governor-elect Hinojosa's call for legislative guardrails. We also expect Texas to remain a business-friendly state and recognize the benefits that data centers bring to local communities through tax revenue and jobs. As we discussed earlier in the context of Beacon Point and River Bend, our focus is on building responsibly with local communities in mind, a commitment that we believe aligns with Governor Hinojosa's objectives. We do not believe the results change the outlook for Beacon Point, which, as a late-stage project, is well-positioned. For Beacon Point Phase 1, our baseload designation remains subject to Governor Abbott's audit.
We believe we are well-placed in that process and anticipate an unconditional classification on December 17, which will provide a clear path to energizing the full gigawatt at Beacon Point, with 700 MW expected in the first quarter of 2027. With that, happy to take questions.
At this time, in order to ask a question, press star, then the number one on your telephone keypad. If you would like to withdraw your question, press star one again. Thank you. Your first question comes from the line of Ryan Gravett with UBS. Please go ahead.
Great. Thank you. Asher, I did want to just follow up on your closing remarks there, just given the election outcome last night. Anything you can share in terms of various other permits required for both phase one and phase two at Beacon Point? Then maybe one for Sean. Anything you can share in terms of expected timing for project financing at phase two of Beacon Point, and if the election outcome changes that timeline at all? Thank you.
Thanks for the question, Ryan. In regards to the build, we're primarily focused on the December 17 timeline. When we think about TCEQ and any permits in regards to diesel generators, there are paths to be able to design around that and use the BESS system as an alternative backup source. Then those are the kind of the primary focuses for us as we continue to progress and develop the project. Sean, I'll pass it to you.
Yeah. The sort of timing that is right in front of us does not really change. We expect to hear back from the audit results on December 10 and then have those be finalized on December 17. Those remove largely the sort of remaining impediments ahead of us to get going on the financing for Beacon Point 2. There are a lot of different ways that we can see to finance that. Obviously, we have addressed and tapped the investment-grade bond market, and we think there are a number of other avenues around that that exist too. We feel really confident about the ability to get that done. In our minds, the path is the same as it was before, and so we look forward to executing on that.
Got it. Thanks, guys.
Your next question comes from the line of Eric Luebchow with Wells Fargo. Please go ahead.
Great. Thanks for the question. I am curious, given everything going on in terms of higher labor costs, more upfront investment in the sites in terms of utility deposits, what are you seeing in terms of your build costs, not just for Beacon Point and River Bend, but as you look at your power pipeline today? Are you seeing build costs move into the low to mid-teens on a dollar per critical load megawatt basis? Do you think that you have enough pricing power to kind of maintain or even improve your development yields as you build out the next phase of your pipeline? Thank you.
Appreciate the question, Eric. I'll answer that in three parts. Part one is origination, part two is the development and EPC, and part three is the commercialization and lease rates. For part one, in regards to origination, you're right. A lot of utilities now ask for more upfront collateral, CIAC payments to make sure that the infrastructure upgrades or energy supply that's committed to doesn't affect ratepayers. We've been supportive of that since day one, and that's the same structure that we've had at River Bend and Beacon Point in paying for those CIAC upgrades and any collateral that's necessary. As we continue to develop, our approach remains similar, which is we don't put large development capital at risk in the pursuit of commercialization.
We stay disciplined in the origination process, which makes it more difficult to originate, but we found opportunities where we are able to do so, able to bring customers along for the process with us, and be able to continue to develop the sites that we have conviction in. As our balance sheet continues to grow, we obviously will be working on more sites across our pipeline. As you see, this quarter, we had another 1,300 MW of sites go from energy under diligence to energy under exclusivity, which means we're diving deeper into those sites, we're spending more resources in developing those projects. We really think about it as a percentage of overall development capital as we continue to scale that we can increase to develop those sites.
But we are spending, call it, hundreds of thousands or single-digit millions upfront in these developments, and not tens of millions or hundreds of millions of dollars. Really, when the collateral requirements come, we're deeply engaged with the end user during those moments in time. In regards to EPC, everyone has heard labor rates have gone up, you have supply chain rates that have changed, and as we've continued to build over the larger of a decade here, the focus has been, how can we control our destiny? We can't control our macro conditions of overall capital markets, of overall cost of labor, equipment, et cetera. What we can control is, how do we design the infrastructure to be able to drive additional efficiency to try to offset those changes and those increases that you had mentioned?
If we can ruthlessly focus on learning from the builds that we have today, designing from first principles, driving efficiencies in the design, which ultimately drive efficiencies in the supply chain and the equipment we need to purchase, which ultimately drives efficiencies in the construction and the amount of hours we need to construct a data center, those are things that are in our control and that we have conviction and confidence that we continue to build towards. So, that is our offset against these macro conditions that you mentioned and I chatted about earlier in my prepared remarks on how we want to build a scalable, repeatable model, like the Raptor photos of the Tesla engine that I'm sure many people have seen during iteration one, two, and three, or SpaceX, versus kind of a bespoke build every single time that is expensive and costly and time-intensive.
And then the last one is in terms of lease rates. We've always approached the market from a perspective of signing leases that we believe are fair and kind of middle of the fairway market from a lease rate perspective, and our shareholders and our investors have been happy with the structuring and the returns on those deals. So the way that we think about it is, if market costs on average across the industry go up, therefore lease rates would move in proportionally with that because many folks work on a more yield on cost basis. So, we'll charge market-based rents, meanwhile focus on driving ruthless efficiency within our execution EPC stack to construct and hold costs on construction, which could in turn actually increase the yield on the projects that we're developing if revenues go up and we're able to maintain costs.
Thanks, Asher.
Your next question comes from the line of Bryan Baratian with ROTH Capital Partners. Please go ahead.
Hey, guys. Thanks for taking my questions. I have two. One is, as we, in the face of the election, and looking to what I think is one of the most interesting things is River Bend expansion, where you have 1 GW of IT capacity to expand there. How can we think about what the structure looks like with regard to ultimate counterparty? I know at the River Bend site day, the head of compute at Anthropic was there, and it all seemed very positive. But would we expect to see, similar to your other leases, the same level of investment-grade equity support that was both across River Bend and Beacon Point with different tenants? Or, how are you thinking about structuring that?
Is there a second one? Or do you want me-
Oh, the second-
to answer first?
Oh, yeah. You can start with the first, and then I'll do the second. It's separate.
Okay, sounds great. We are very excited for River Bend. We have been for quite some time and have shared so with the market. There has been a good amount of work that has been done. We believe that there is actually more than 1 GW of gas available on the pipelines that run through the property, and have been working towards being able to actualize and realize the continued expansion of that campus. When we look at the end commercialization, obviously I cannot speak to any deals that we have not announced yet, but we continue to stay focused on the credit quality of our end offtake relative to the investment and the capital we ultimately need to raise.
We are pretty open. I was on a call with a non-AI lab, non-investment grade counterparty yesterday, someone we admire a lot in the business that they have created, and we were open.
We said, "Look, it is very hard for us to tell our shareholders and our board to go and build infrastructure for folks that do not have the credit that supports the ability to bring on the financing that we have brought on historically to be able to build and execute these projects at scale." I do not want to speak to any deals that are not announced yet, Bryan, but overall, maintaining balance sheet, maintaining credit profile of our cash flows continues to be a core focus of the business and ones that shareholders have asked us for as well.
Awesome. Thanks. The next one is, as we think about the existing footprint of opportunities that you have, I know American Bitcoin is running north of 400 MW of capacity. I guess in the face of the election that happened today, how are you, or if you can provide any color on, thinking about potential brownfield opportunities as it relates to the existing footprint there? Or is the focus just going to continue being on greenfield development?
When you look at our pipeline that we have announced, the development pipeline is almost 8 GW, with a total of about 8.6 GW in terms of our overall pipeline overview. That is one out of four buckets of the ability to commercialize and realize additional TCV. That bucket is our greenfield origination bucket. The second one we have is the behind-the-meter generation bucket, which I chatted about more so in your first question in River Bend and other sites that we have. The third one is M&A, and through our corporate development team, there continues to be very interesting opportunities and growing opportunities in a world where capital becomes more constrained and flows to proven operators. The fourth is brownfield opportunities as well in existing campuses.
We continue to look at all four of those, continue to originate across all four, which gives us strong conviction that outside of the publicly disclosed pipeline, we have three other pipelines with dedicated teams looking at opportunities there.
All right. Thanks.
Thank you. Appreciate it.
Your next question comes from the line of George Sutton with Craig-Hallum. Please go ahead.
Thank you. Asher, I found it interesting that your energy capacity under diligence came down, but your capacity under exclusivity went up. Can you just give us a sense of, that just sounds like a more focused way to go to market. Is that the right way to read that?
That's exactly right. The way to think about our pipeline is a funnel. You start top of funnel more wide, and as you get deeper and deeper in, more resources get allocated, more time gets committed to sites, closing the aperture. At every stage, the requirements get higher and higher as we continue developing those sites. We have clear guardrails before projects pass from one stage to another, and we dedicate additional resources to it. So you see a transition from capacity and due diligence to exclusivity through quarter over quarter.
Just one quick question on when we watched you build Vega, that was just simply a better mousetrap at the time, really shortening delivery times and lowering costs. When you look at the application of these kinds of principles to the AI development, can you just give us a sense of how significant can those deltas be in your view, and maybe looking out a couple years?
Very significant. We've done more than a couple of tours in recent months at Vega, sharing our design philosophy and abilities. It's actually pretty interesting because before we started building River Bend and Beacon Point, we had self-engineered, designed a lot of our own infrastructure and GC'd all of it. So Vega, if you go to that campus, you not only see the first principles design methodology, but you see the components in which we designed and contract manufactured everything from the pipe skids to the PDUs, the switch gears, and so forth. As we entered the AI space, the big thing that we were building was credibility for end-user tenants to trust that we could build this infrastructure stack. So we hired some of the largest GC equipment suppliers in the world to help build that credibility and help execute.
As we've executed for these end users, we've built that credibility, and in those discussions, we started talking about our origination, where we came from, the capabilities that we really haven't been able to tap into on these projects because we haven't been able to use first principles design from the ground up, how this infrastructure stack would work. That's opened a slew of conversations as people think about this capacity, as people think about this capacity on Earth and space, how do you drive more efficiency, more speed. There's a next chapter there, George, that I alluded to in my prepared remarks, but we believe the skill sets and the philosophy that we built in a world where we were able to build infrastructure without constraints will be coming to light as well.
Super. Thank you.
Your next question comes from the line of Ben Sommers with BTIG. Please go ahead.
Hey. Yeah, good morning, and thank you for taking my question. I wanted to just ask a bit about the capacity that you guys moved under exclusivity. It says that you're looking across multiple regions there, so anything you want to highlight of particular regions of interest as we expand the power portfolio?
Yeah. When you look at the overall pipeline, one big ask that we received from investors was understanding the pipeline and the makeup of that pipeline. We shared the megawatt numbers, but we didn't share how many sites those were. We didn't share what regions they were made up of. As you look through the disclosure this quarter, we shared the average size per site. We also shared the makeup of what ISOs those sites are, and I think for most people, they'll see that we're very diversified. The thing that causes that diversification is we actually have teams dedicated to different regions to specialize in those regions and to dive deep within them as well. As we continue to scale, we have our greenfield origination team, we have our corporate development team, so we're looking at projects across the board.
As we think through the election cycle yesterday as well, diversification, I think, is increasingly important to be able to navigate through any policy or market volatility that may come up, and we feel confident in that diversification today.
Super helpful. Thank you for taking my question.
Of course. Thank you.
Next question comes from the line of John Todaro with Needham & Company. Please go ahead.
Hey, guys. Thanks for taking my question, and congrats on a quarter down here. I guess just going back to Texas and the election, and your comments around ERCOT and the December approval schedule, just wondering if you could frame up the confidence in that. My read would be maybe it gets pushed out and you get new directives from a governor, but you guys are obviously closer in Texas and with those regulators. So just wondering on the framing of ERCOT confidence for the December approvals.
Going into the elections, we spoke to a lot of our consultants and lobbyists in regards to the different outcomes of the election and how it was effective. I think the thing that is really important is from day one, we've focused on building responsibly. And so even when the audit process came out, we were confident there because we had already answered the questions from the PUC that came out that were voluntary. Similarly here, we believe the mindset is not, don't build net new investments into this state, but more so, how do we build responsibly? And so based on the current progress and being a very mature project that's financed, and the way in which we're building, we feel confident in working with the governor-elect in anything that she needs in terms of continuing to progress and invest into the state.
Understood. Thank you, Asher.
Thank you.
Your next question comes from the line of Patrick Moley with Piper Sandler. Please go ahead.
Yes. Thanks for taking the question. I just had one on the buyback, Asher. You said that you felt the shares didn't fully reflect contracted revenue and controlled power. Could you maybe just share how you frame intrinsic value internally? Is that an NOI multiple? Is it sum of the parts or something else? At what point do you think you could look to fund repurchases with cash or the revolver rather than Bitcoin, given that the stack has- I think it sits a little under $700 million now. Thanks.
Thanks for the question, Patrick. It's really a triangulation on intrinsic value because many people look at the company and value it differently. The first methodology is we look at the sum of the parts. What are the two, three contracts that we've announced across the two projects? So that's River Bend, Beacon 1, Beacon 2. What is both the net present value of those contracts, so doing the DCF, but also how do we think about the cap rates in different market scenarios that we'd be able to sell those projects and what they should be trading at? So that's existing projects. Then we have conviction over the four funnels that we have, whether that be the greenfield funnel that we disclosed publicly, the behind-the-meter funnel, the corporate development funnel, the new opportunities we're looking at, and the brownfield funnel as well.
That creates the kind of growth of where we believe the company can go, and we apply a discount factor to different projects at different stages. If we look at, I mean, where we are trading today, we are not even trading at the value of all three agreements, needless to say, even two of our contracts. We believe that we are trading, I mean, today even more so, compared to when we were in the 70s, a discount to the value of the existing agreements if they just come online and start operating on these triple net structures. Then we have all the growth that we have in plan, and put aside gigawatts that we have in our pipeline, nearing 10 GW of opportunities, even if we were to execute another one or two sites, that would increase the value of the company substantially.
That is how we have thought about intrinsic value of the company. As we think about kind of what stage we are at in the market and how do we think about shareholder buybacks, the market has obviously had some negativity over the recent weeks. Even though we may be trading under intrinsic value, we are focused now to make sure that we maintain capital preservation and optionality, because there might be interesting buying opportunities that come up that never came up before because things were too expensive. We do not plan to continue to buy aggressively as we want to reserve that cash. In our revolver, we also do not plan to use the revolver for share buybacks.
We thought it was a really nice way to respond to investors over the last two years, which the main question we had around balance sheet was, how long do you plan on holding Bitcoin for? You are an AI company now, you do not get value for your Bitcoin at all. Why still hold it on the balance sheet? As we looked at where the markets were relative to our share price and Bitcoin price, frankly, we had a lot more conviction in the appreciation of our stock relative to where Bitcoin price was as well. We swapped exposure from Bitcoin to Hut 8 shares and initiated those buybacks.
All right, great. Thanks, Asher.
Your next question comes from the line of Logan Hennen with Northland Capital Markets. Please go ahead.
Hey. Morning, guys. Thanks for taking our question. Just one from us today. Can you provide some additional color around the new 230 MW acquired in West Texas in regard to how you're thinking about that opportunity, and maybe clarify where that site sits in Batch Zero and the remaining permits and approvals needed there? Thank you.
Happy to do so. We were a part of a bankruptcy. It was kind of a deja vu in 2022. We were a part of many bankruptcies where we acquired assets during the downturn in the crypto market, and we were able to acquire many infrastructure projects, whether that be full acquisitions, taking on managed services agreements, or being plan sponsors to spin out new companies like Ionic. It was a very informative journey in our lives, and when we were a part of this bankruptcy auction, it was a little bit of a deja vu back then. We grew almost 10x during that period of time as a company. Building during market downturns, we see as really interesting opportunities to grow because things are really cheap. It's never as ideal to build when things are expensive.
It's always much more exciting to build at a lower cost basis. I think this was the first opportunity in the market that allowed for that. We've been patient. We felt like at some point, the opportunities like this would come up, and one of them did come up, so we pursued that opportunity. Those two sites are located within driving distance of our Salt Creek campus in West Texas. You have sites there and power for the Bitcoin mining sites that were there that are energized right now, or that are energized. They don't have miners there that were hashing because through the bankruptcy, those hosting clients left. We visited those sites last week, and we are working through the closing conditions now and look to close by end of year.
Your next question comes from the line of Richard Choe with JP Morgan. Please go ahead.
Hi, Asher. I just wanted to ask, as you've gone through this deal with the leading AI company, how has the change orders and the engineering changed, and how has Hut been able to maximize your ability to address those needs? Because I think the ability to build these large-scale AI data centers is underappreciated, and just wondering-
Yeah.
what challenges that you've been faced with and how you've handled them.
Appreciate the question, Richard. I'll answer that question from two mindsets, current status quo of the projects we're building and net new approach towards building that I've talked about a bit on this Q&A and the prepared remarks. I'll go through a little bit of a flowchart of the different areas in which you move through them. The first thing is a change order comes in. That change order may be driven by insurance need, may be driven by customer needs, whatever it may be. If you look at a site like River Bend, there are a lot of counterparties involved, right? Our end tenant is Fluidstack, then you have Anthropic as the end user. You have Broadcom and Google who are part of the credit profile. We have Jacobs as our GC, their subcontractor is Vertiv.
There's a lot of parties in the mix in here. Any change order has to run through the system and through all these various parties in many instances. The first question is, when a change order comes in, is that a change order that is necessary or not, is the first question. There's discussion around what is the ultimate thing that is trying to be solved for, and is this the best way to solve for it or not? The easy thing is say, "Okay, you gave me a tenant change order. This is the time impact, this is the schedule impact that we received from Jacobs. Here you go. Thank you. Do you want to pay it or not?" We think that that starts to have a very big creep situation, where things creep up because you just take it for gospel.
We start with challenging the first principle, working together across all of the partners and saying, "What is the ultimate thing that we're solving for, and is there a more efficient way to do it without impacting schedule and cost?" Because ultimately, schedule and cost, if we didn't do that, just get passed through to the tenant, but we want to drive infrastructure coming online as quickly as possible. Then, if we get to a place where a change order is necessary, whether it be a change in requirements, change in design, whatever it may be that causes that, then the question is, what can we do to be able to minimize the impact of that change order?
Once you get to that final answer, everyone's aligned on this is the change order that's necessary and this is the impact that it has, then there's a slew of questions that arise. The contracts currently have it so change orders need to be paid back in cash, reimbursed in cash. The schedule impact applies to the lease, kind of pari passu with wherever the impact is, and you get a fee on top of that for managing the process. There's obviously increasing questions from end users, which is, "Hey, can you rentalize this? Rather than it being just a CapEx reimbursement that we pay for, can you rentalize this or not?" I think when you think about that, there's really multiple ways to think about that.
One is there ample ability within the budget to be able to rentalize it, that is within the financing guardrails that we have built towards and the contingencies that we've put in place? If so, then great, you get more yield on the capital stack that you have. The second is, if that doesn't exist, can you use external capital or additional funding in order to do so? We've seen that in other parts of the market as well. If you look at it from an individual investment, depending on what yield on cost is agreed to on the rentalization, it could be really interesting depending on your balance sheet position, your willingness to take on that return on investment. Obviously, right now, we're investing into these projects and have pretty good structures, LTCs, and so forth, and so that bar is pretty high.
But overall, that's how we would think about it in bucket one, which is the current regime with many counterparties and really these bespoke, unique solutions. And there's a lot of engineers involved, and people have different opinions and drive, and that causes design changes. The second bucket is going back to first principles, Vega, and so forth, which is, what is the thing that we're trying to solve? Which is bring on reliable compute at speed and at scale. And so, there is a second camp in which we've been having a lot of discussions around, which is, can we get aligned from a leadership perspective within Hut 8, our tenant end users, et cetera, that we align on the thing that we need to build?
And if we're able to build it really quickly, we don't allow for that many changes along the way, because those are the things that really slow down and kill a project. And in that scenario, you're building much more like a SpaceX, where you're building at speed, you don't have as many cooks in the kitchen making decisions, and you have a set thing that everyone's aligned on upfront, and then you sprint at getting that infrastructure stack online. And so, Richard, I hope that was helpful. I'm walking through status quo of industry, and then how do you actually bring those in so then you're not on calls talking about those impacts. And it requires a certain type of partner philosophy and conviction and confidence in the ability to build with first principles.
No, that's great. And honestly, I think the problem is people don't realize how this is a problem at scale that you need to engineer for, and everyone just thinks it's easy to build, and it shows the level of detail that is needed to deliver this amount of capacity at scale.
I'll give you an example, if we go a little bit more granular. We had a request to change some of the lineups because there was a different chipset and design that was going in. And at first, when we asked for those requests, the thing that came back was the impact in schedule and cost that we believed was too high. And so, we really dove deep with our equipment vendors and GCs, and we said, "What is driving this?" Because if everything's passed through, as a developer, we can just pass it through to the end user and say, "You asked me for a change order.
This is what we got back, and this is the change." As we think about the importance of trying to bring on compute quickly, we cannot do this across all the board, but the big reason was having to go and change basically the cable trays that have to pull more cable across the facility and the SmartRuns to be able to get them into the actual TPUs. We looked at it, we said, "Okay, why do we have to actually increase the size? Why do we have to go re-engineer all of those skids? Why do we have to increase the sizing? Is there a way to do that in the field so we are able to save on schedule and cost?
We were able to reduce that cost impact by over 90% by just challenging and focusing on how do we engineer this in a better way. Richard, many of those situations come across alongside large builds, and being able to navigate and manage through those with your end users and all of your partners is critical.
Great detail. Thank you.
Your next question comes from the line of Chris Brendler with Rosenblatt Securities. Please go ahead.
Hi. Thanks, good morning. Thanks for taking my question. I just wanted to see if there was any way to think about Louisiana versus Texas, given the election results. No change at the governor in Louisiana. I know that seemed like a very supportive community when we were down there last month. How should we think about your future expansion plans in relation to the election results?
For the folks that had the opportunity to come to our River Bend first look ceremony, the governor was there, the local community was there. We're very proud of the relationships that we've built and the support that we've garnered throughout the state, and the adoption we've had in Hut 8's ability to build responsibly. We're big fans of Louisiana. We were before people even focused and looked at Louisiana. We were laughed out of rooms two, three years ago when we talked about Louisiana. There's a lot more focus today, but we continue to double down and focus on the state and are very proud of the work that we've done, and are excited to continue to invest and continue growth in the state as well.
Great. Then, just a quick follow-up on the buyback/capital needs question. Trying to understand the balancing act there. How do you view the current Bitcoin balance in relation to cash needs? It just feels like not quite the right time to me to be leading into a buyback. So, I'd love a little more color on how you're thinking about your cash position over the near term, given market conditions. Thanks.
Really appreciate the question. I'll start with the buyback itself and talk about our liquidity position. So, in regards to the buyback, we've increased the buyback not necessarily to buy immediately, but to hold that optionality. When we first announced the $250 million buyback, that was almost two years ago or so at this point, and we really only initiated that in the recent quarter. We just sat on that opportunity. Similarly here, expanding the buyback is less around we want to go buy a billion shares immediately. As I mentioned, liquidity and being able to invest in opportunistic moments is critically important, and especially in these market conditions. So, we don't expect to go and immediately go buy shares off the back of this increase, but rather just leave it on the shelf as an additional tool that we have.
Our overall liquidity, if we think about on the parent level and across the projects, is over $7.7 billion. That is across our Bitcoin, the cash we have, the restricted cash we have on the project level that finance those projects, and the revolver that we put in place as well, and the availability along that revolver. We are excited and confident with that access to that capital that we have on our balance sheet and our bank accounts. We also have zero recourse debt at this point on our balance sheet as well. The increase of the share buyback is to have on the shelf and not to implement immediately, because we believe that cash will be king as we navigate through this current volatility.
Awesome. Thanks so much, and congrats on all the progress.
Thank you.
Your next question comes from the line of Nick Giles with B. Riley Securities. Please go ahead.
Yeah, thanks very much. Maybe just a question for Sean. Curious on how the way you are approaching structuring any future debt deals has evolved since your last raise, or can you just give us a sense for some of the alternatives you are exploring or any kind of features of a debt offering that you might be considering today that were not as logical, say, six months ago? Thanks.
Yeah, thanks, Nick. Appreciate the question. So, look, I do not think we actually view how we approach financing any differently from the way we did the last two. We start with a clean slate every single time. We say, "What is the asset we have? What are its attributes? And what do we think is the financing that it deserves?" We obviously also take a look at the market environment. How are different markets performing? Where is their most depth of capital? Where do we see the most liquidity and best pricing? And then we also think about what is going to increase the free cash flow yield of these projects on a risk-adjusted basis, as we talked about in the prepared remarks. So, I do not think the approach actually changes at all. It is like everything we do at the company from a first principles perspective.
We have been preparing for the financing for Beacon Point Phase 2, which I think is kind of what you are alluding to, for a while. And we have, I think, really great solutions that we are going to bring to bear. Some of those might involve things we have done before, but they do not necessarily. But we have been working pretty tirelessly to make sure that we have a great path. I think we got a lot of kudos for the prior financings we have done, and we are not resting on our laurels. We have a very, I think, clear path on what we want to do, how we want to finance, and that is going to A couple things are not going to change.
One, recourse to the parent, and two, we are going to try to minimize as much refinancing risk as we can. So, again, approach does not change.
I think some of the instruments and tools we use might. But we will keep that for the day when we finally do announce our financings, because we do not want to obviously, forerun any transactions.
Appreciate that clarification, Sean. I'll turn it over, but continue best of luck.
Thanks, Nick.
Your next question comes from the line of Dillon Heslin with Lucid Capital Markets. Please go ahead.
Hey, thanks for taking my question. On the Beacon Point Phase 2 lease amendment from 15 years to 17 years, could you just provide a bit more color on how you're looking at that? It seems like you mentioned higher financing costs. Is that an interest rate look-through, or is that a potential combination of both the CapEx flowing into a higher bond stub that you need and thus interest rates? Does that still mean that two-year amendment still doesn't eat into any of the extensions, correct?
That is right. That is exactly right. It does not eat into any of the extensions. And really, it was a holistic discussion over the site, and I think there is a big question that folks have asked, which is, if there are changes in market conditions, how do the tenants respond to that? And at the end of the day, the data center costs are still relatively small to the overall compute economics and the value of those economics. So for us, I think this is a signal and a sign that, whether it be through tenure, rates, et cetera, that tenants understand market conditions and are willing to support to make sure infrastructure gets built.
Great. Thank you.
Your next question comes from the line of Brett Knoblauch with Cantor Fitzgerald. Please go ahead.
Hi, guys. Thanks for taking my question on this wonderful morning. I just want to hit on Texas one more time. Obviously, kind of the outcome we were not expecting. When we read through maybe the incoming governor's previous stance on data centers, it does not feel like she is actually super against them, and it is not a hard line, no data centers. It seems like she is in line with the rest of the country and wants data center operators to pay for grid infrastructure, water infrastructure, to save ratepayers, local communities to have a say in what is happening. How different do you view the Texas landscape today versus yesterday? Is it more of the same? Should we assume that most tenants now are going to be looking outside of Texas? Just any additional color would be helpful. Really appreciate it.
I agree with you, Brett. We believe that this infrastructure is an American initiative. It is not right or left. And whichever side you're on, making sure that citizens don't get impacted in their day-to-day lives in terms of their energy, their water, their status quo of their lifestyle is critical. The way that we've been building has been with that mindset and that approach, right? If you look across the states in our development pipeline, there are states before last night that had Democratic leaders within those states, and we've continued to explore and continued to push forward on those developments of those projects because we believe we have a program that's focused on building responsibly.
As a result, in Texas, we don't actually think this changes much of what the governor-elect is looking to do because we're very supportive to make sure that citizens are protected while America is building this critical infrastructure. We're excited to work alongside her and her administration to make sure that Texas can lead the front as well in terms of building this infrastructure and having these investments go into the state responsibly. My perspective is similar to yours, Brett, and we continue to focus on that foundation regardless of people have asked for it or not.
Really appreciate it. Thank you, guys.
Thank you.
That concludes our question and answer session. I will now turn the call back to Asher Genoot for closing remarks.
As we have built this business, we've always understood that periods of rapid infrastructure growth come with volatility, and there will inevitably be peaks and troughs driven by both broader macroeconomic conditions, but also industry dynamics. What has always mattered to us is building a business that can not only withstand these cycles, but also continues to grow and execute through them. That is why we've been so deliberate in strengthening our balance sheet, structuring our business thoughtfully, and building a foundation designed for long-term success. As I shared last quarter, our excitement and conviction in the business we are building remains as strong as ever. We know there will be challenges along the way, but we're confident in our ability to navigate through them, overcome them, and emerge stronger.
I reminded our team in our last town hall of Warren Buffett's famous observation that the market is a voting machine in the short term and a weighing machine in the long term. We can't control short-term market conditions, but we can control how we execute, how we allocate capital, and how we build this company. Our focus remains on what has always mattered most, executing on our commitments, delivering for our customers, and building a business that can compound value through every market cycle. We believe the opportunity ahead of us is extraordinary, and we remain as committed as ever to realizing it. Thank you all for being with us today.
Ladies and gentlemen, that concludes today's call. Thank you all for joining. You may now disconnect.