Earnings Call Transcript - KEEL Q1 2025

Operator, Operator

Thank you for standing by, and welcome to Bitfarms’ First Quarter 2025 Earnings Conference Call. At this time, all participants are in a listen-only mode. After the speaker presentation, there will be a question-and-answer session. Operator provided instructions. I would now like to hand the call over to Tracy Krumme, SVP, Investor Relations and Corporate Communications. Please go ahead.

Tracy Krumme, SVP, Investor Relations and Corporate Communications

Thank you. And welcome to Bitfarms’ first quarter 2025 conference call. With me on the call today is Ben Gagnon, Chief Executive Officer and Director; and Jeff Lucas, Chief Financial Officer. Before we begin, please note this call is being webcast with an accompanying slide presentation. Today's press release and our presentation can be accessed on our website, bitfarms.com, under the Investor section. Turning to Slide 2. I'd like to remind everyone that certain forward-looking statements will be made during the call, and that future results could differ from those implied in this statement. The forward-looking information is based on certain assumptions and is subject to risks and uncertainties, and I invite you to consult Bitfarms’ MD&A for a complete list. Please note that references will be made to certain measures not recognized under IFRS and therefore may not be comparable to similar measures presented by other companies. We invite listeners to refer to today's press release and our MD&A for definitions of the aforementioned non-IFRS measures and their reconciliations to IFRS measures. Please note that all financial references are denominated in U.S. dollars, unless otherwise noted. I’d also like to remind everyone that we will be at the following upcoming conferences: Consensus ’25, which takes place today through May 16 in Toronto; the AIM Summit in London from May 19 to 20; Bitcoin 2025 Conference in Las Vegas from May 27 to May 29; HPE Discover in Las Vegas from June 23 to June 26; and Northland's Virtual Growth Conference on June 25. There will be opportunities for in-person one-on-one investor meetings around these conferences. So if interested, please reach out to investors at bitfarms.com. And now, turning to Slide 3, it is my pleasure to turn the call over to Ben Gagnon, Chief Executive Officer and Director. Ben, please go ahead.

Ben Gagnon, Chief Executive Officer and Director

Thanks, Tracy. Good morning, and welcome to Bitfarms’ Q1 2025 earnings presentation. I'm thrilled to address our esteemed audience of investors, analysts and stakeholders who have joined us today. Your support and engagement are invaluable as we navigate one of the most transformative periods in Bitfarms’ history. Turning to Slide 4. The global demand for compute power is surging, driven by AI innovation, cloud computing and data intensive applications. Bitfarms is fully equipped to meet this insatiable demand with our robust North American energy portfolio and operational expertise. A recent McKinsey report cited that by 2030, companies across the compute value chain need to invest a staggering $5.2 trillion into data centers to meet worldwide demand for AI alone. This is based on a projected 156 gigawatts of AI-related data center capacity demand by 2030, 125 of which will be added in the next five years. In five years the power needs for HPC and AI will be almost 10 times the power needs of Bitcoin today. The bottleneck on the growth is not chips but power, and this is where Bitfarms comes into play. We are no longer solely a Bitcoin mining company; we are evolving into a leading North American energy and compute infrastructure company. This strategic pivot leverages our core competencies and energy portfolio, positioning us at the forefront of the exciting high growth sectors of high-performance computing, artificial intelligence and energy infrastructure, while maintaining cost effective and efficient upside to rising Bitcoin prices. Today, we’ll guide you through our solid progress in Q1 with a focus on two key priorities for 2025: continued U.S. expansion and advancing our HPC and AI business. By focusing on these priorities, we aim to drive long-term shareholder value, capitalize on macro tailwinds in energy and compute markets, and establish Bitfarms as a leader in powering tomorrow’s economy. Turning to Slide 5. In Q1, we made extraordinary progress in rebalancing our portfolio toward the United States and HPC and AI infrastructure, executing two transformative transactions that set the stage for our future growth. First, we acquired strategic U.S. energy campuses and power generation facilities giving us immediate HPC development opportunities and a robust multi-year growth pipeline in Pennsylvania. With these two flagship campuses each boasting a multiyear pathway to nearly 500 megawatts of power capacity in prime locations, we are attracting interest from institutions with a robust appetite for power and compute infrastructure. Second, we strategically divested our Yguazu, Paraguay Bitcoin mining site. This site was purpose built for Bitcoin mining and lacked the fiber and market demand for HPC and AI conversion, making it misaligned with our new HPC and U.S.-centric strategy. The strategic site divestiture was profitable and brought in significant cash proceeds for reinvestment in our U.S. growth initiatives. More importantly, it allowed us to avoid hundreds of millions of dollars in capital expenditures needed to complete and equip the site for Bitcoin mining, preserving our balance sheet strength for higher value opportunities and keeping our capital, time and efforts focused on the U.S. and HPC. The rapid and simultaneous execution of these transactions demonstrates Bitfarms’ agility and forward-thinking approach towards managing and developing our energy and compute infrastructure portfolio. We’re not just planning for the future, we’re building it. To support this pivot, we’ve restructured our organization, streamlining operations and aligning teams for HPC and AI development in the U.S. We’ve onboarded top-tier executives with expertise in HPC and AI and construction management, enabling us to collaborate at an accelerated pace with our world-class advisors, WWT and ASG. These advisors have brought extensive resources and expertise, helping us assess, design, and co-market our HPC assets to prospective clients, significantly collapsing development timelines and ensuring better outcomes. Lastly, on the Bitcoin mining front, we grew our Exahash Under Management over 50% in the first quarter to 19.5 exahash. We are now incredibly well positioned to take advantage of rising Bitcoin prices with almost no planned Bitcoin mining CapEx remaining. Importantly, with fewer than 300 miners left to clear U.S. customs and those 300 scheduled to clear this week, we have practically no foreseeable tariff risks on our miners. Over 94% of our purchased miners are now installed and with the final install of remaining miners occurring this quarter, we’ve effectively completed our planned Bitcoin mining growth initiatives. Through this fleet upgrade, we have gained market share and reduced our operating cost per terahash dramatically. Importantly, with no material CapEx planned for Bitcoin mining, the majority of our current 2025 CapEx investments will focus on electrical infrastructure, namely substations, transmission lines, and civil works. The capstone of our Q1 achievements actually occurred on April 1, when we secured up to $300 million from a division of Macquarie Group, one of the world’s largest infrastructure investors. The deal validates our HPC and AI development thesis and provides the capital to begin development of our Panther Creek campus, which I’ll discuss shortly. The initial $50 million tranche is secured by Bitfarms’ assets, with an additional $250 million that will become available as we achieve specific development milestones. This partnership not only strengthens our financial position, but also opens doors to other top-tier collaborators, enhancing our credibility and outcomes in the HPC and AI market. Turning to Slide 6. The foundation we built in Q1 was what we needed to do as a company to facilitate our pivot to HPC and the U.S., and I am proud to say that we accomplished in Q1 exactly what we sought out to get done. Now with this foundation complete and with the right people, advisors and assets in place, we are focused on developing our HPC business in a systematic and scalable fashion. Our HPC and AI business is being built on a structured, repeatable process designed for scalability and efficiency across our portfolio of data centers. For every site we seek to convert into HPC, we will develop the master plans while pursuing site-specific strategic financing partners or structures to finance the infrastructure development in an accretive fashion to our shareholders. Afterwards, construction of infrastructure and the customer acquisition process will both ramp simultaneously in order to accelerate timelines. This repeatable approach ensures we can streamline workflows and replicate our success and learnings at Panther Creek across other sites in our portfolio while maintaining consistency, optimizing resources, and wherever possible collapsing timelines. I would now like to take a moment to zoom in on Panther Creek, where we have secured initial financing and are making great progress on the master site planning. Turning to Slide 7. With a potential capacity of nearly 500 megawatts, supported by multiple power sources, ample fiber access, and strategically located near major metropolitan areas and existing data center clusters, Panther Creek is going to be our first HPC campus. We’ve made significant progress on Panther Creek’s development since we secured the site in March. On this slide, we have some renders of one of the conceptual data center plans for Panther Creek that are coming out of the design and engineering of the site master plan. So far, we’ve created preliminary site map plans, 3D models, and test fits for various data center designs, allowing us to optimize the campus layout across numerous phases of development and designs. Phase 1 and Phase 2 substation design and engineering are nearly complete, and we’ve also conducted site visits with leading manufacturers and suppliers of modular data center infrastructure, exploring innovative solutions to accelerate deployment and reduce costs as we build for the HPC needs of the future. Further engineering, design, and planning are underway, with the goal of finalizing the master plan for Panther Creek in the second quarter. Once complete, we’ll break ground in the second half of 2025, initiating construction of transmission lines, securing key substation equipment, and beginning civil works to prepare the land for active construction. Simultaneously, we’ll be ramping up customer acquisition efforts, engaging prospective clients to refine the infrastructure based on their specific needs. This dual-track approach ensures we’re building a market-ready facility that maximizes value without wasting any time. Turning to Slide 8. I’m excited to share that the feasibility assessments from WWT and ASG have validated the suitability of all of our U.S. sites, Panther Creek, Scrubgrass, Sharon and Washington state for HPC and AI conversion. These sites boast the critical attributes needed for high performance data centers, immediate power capacity, ample land, robust fiber networks, proximity to major metropolitan areas, and positioning in high-demand compute markets. For example, Panther Creek, Scrubgrass and Sharon in Pennsylvania benefit from their location in the PJM Interconnection, one of the largest and most reliable power markets in the U.S. with access to multiple power sources for enhanced reliability. In Washington, our site is strategically positioned in a region known for its stable low-cost hydropower below $0.03 per kilowatt hour and the existing data center cluster that has developed around it, making it an ideal location for HPC and AI development. These assessments confirm that our U.S. portfolio is not just viable but exceptionally well suited for conversion. Each site is ready to support the infrastructure demands of AI workloads, from power intensive GPU clusters to high speed data connectivity. Our ultimate objective will be to convert all U.S. sites to HPC and AI over time, starting with Panther Creek as our flagship campus. This validation strengthens our confidence in our strategy and positions Bitfarms to capture a significant share of the rapidly growing HPC/AI market. Importantly, the work at Panther Creek also serves as a blueprint for our broader U.S. portfolio. We’re undertaking similar planning and engineering across all our U.S. sites, creating a scalable pipeline of HPC and AI data centers to bring to market. Each campus and site will benefit from the lessons learned at Panther Creek, improving efficiency and reducing development timelines as we expand development across our U.S. portfolio. Turning to Slide 9: As we look to the future, Bitfarms is poised for extraordinary growth in 2025 and beyond. Building off our strong Bitcoin mining experience and foundation, our strategic pivot to HPC and AI infrastructure aligns us with the rampant demand for compute power, driven by AI innovation, cloud computing, and data intensive applications. With a clear path to 1.4 gigawatts power capacity, we’re building an exciting energy and compute portfolio to power tomorrow’s economy. Our energy portfolio has been rebalanced to 70% North American. With all U.S. sites validated for HPC and AI conversion, we are well positioned to maximize the value of our power portfolio through HPC conversion and have the structured development processes, world-class team and advisors to do so rapidly and efficiently. Our Bitcoin mining operations produce free cash flow to support the business and our HPC development and our Bitcoin One program builds off the success we achieved in 2024, delivering cost effective exposure to rising Bitcoin prices and seeking to amplify returns. Lastly, we are doing this with a financial foundation that is stronger than ever. With up to $300 million in financing from a division of Macquarie Group, coupled with strong liquidity, we have the capital to execute our vision accretively with minimal dilution. Bitfarms is not just adapting to the future, we’re building it. For our shareholders, this represents an unparalleled opportunity to invest in a company at the forefront of a transformative industry, with significant upside potential that we believe has yet to be recognized by the market. Thank you for joining us on this exciting journey. I’m confident that our strategic vision, disciplined execution, and strong financial position will deliver exceptional value in the quarters and years ahead. Turning to Slide 10, and with that, I will turn the call over to Jeff for the financial update.

Jeffrey Lucas, Chief Financial Officer

Thanks Ben, and thanks, everyone for joining us this morning. Before we dive into the first quarter numbers, I would like to highlight a few key elements about our financial position: We are well capitalized for 2025 and beyond, and our financing model as we focus on HPC and AI is straightforward. Our Bitcoin mining business remains solid, achieving steady mining margins and providing a consistent cash flow stream to fund our G&A and debt service as we build out Panther Creek. With our miner upgrade substantially complete, we have no plans, nor the need for any substantial miner purchase programs. Our low CapEx mining business is largely de-risked and is well positioned to benefit from any potential Bitcoin upside. And, importantly, the CapEx requirements to grow our HPC and AI business in the near-term are funded with the recent financing secured from Macquarie Group. In short, with steady mining economics, no plans for additional large miner purchases, minimal impact expected from potential tariffs, and the remainder of this year’s capital expenditures funded or with financing in place, we are confident that our financial position will enable us to efficiently and cost effectively grow our U.S. HPC and AI business. Turning now to Slide 11. We are excited to join forces with Macquarie to finance our HPC and AI business cost effectively with minimal dilution. In addition to funding the initial phase of our buildout of Panther Creek, their expertise and their vast experience in HPC and AI infrastructure financing will be integral as we look to further scale the project and expand to other sites within our portfolio. Our highly valued and appreciating North American assets, combined with the higher margin and predictable earnings stream characteristic of HPC and AI, enabled us to secure this attractive debt facility. These funds are dedicated to financing the HPC data center development at our Panther Creek location. Announced in April, the initial tranche of $50 million will cover the soft cost development, with the $250 million project financing tranche available for drawing as we achieve specific development milestones. Turning to Slide 12. Turning now to our first quarter financial performance. With the acquisition of Stronghold, our operations expanded to encompass power generation and hosting. In the first quarter, we earned and received 699 Bitcoin for total revenue of $67 million, up 33% year-over-year. Revenue from our mining activities was $65 million, with the balance of about $2 million from our Volta electrical services subsidiary and from Stronghold’s Bitcoin hosting and electricity generation businesses following our March 14 acquisition date. Our gross mining profit was $28 million, representing a direct mining margin of 43% and an average of $40,100 per Bitcoin mined. Cash G&A was $20 million and included unusual and non-recurring professional services fees of about $2 million in the quarter, primarily for expenses incurred in connection with the Stronghold acquisition and sale of our Yguazu facility to Hive Digital Assets. Operating loss was $32 million in the quarter and included $17 million of impairment charges attributable largely to our Argentina operation and the impact there of higher energy prices and unfavorable foreign exchange rate movements. As a result, net loss for the first quarter was $36 million, or $0.07 per share. Turning to Slide 13: For the first quarter, our Adjusted EBITDA was $15 million, or 23% of revenue. This encompasses primarily our self-mining revenue supplemented by income earned on our hosting activities and our electricity generation. Focusing on first quarter operating performance and per-Bitcoin metrics, gross mining profit was $28 million, or 43% of mining revenue. Our direct mining cost per Bitcoin in the first quarter was $47,800 with our all-in cash cost to mine a Bitcoin at $72,300 compared to revenue per Bitcoin earned of $92,500 resulting in profit per Bitcoin of just over $20,000, for a cash profit contribution from our mining activities of about $14 million. I wish to note that, as we continue to prioritize growing our HPC and AI business, we no longer plan to publish monthly production reports. Instead, we will be providing regular updates on progress with our HPC and AI initiatives. We believe this aligns more closely with the strategic direction of our business. Turning to Slide 14: I will now speak to our liquidity and our anticipated capital needs. As of May 13, we had total liquidity of approximately $150 million comprised of cash and unencumbered BTC. In addition, under the Yguazu sale agreement, we expect to receive from Hive $26 million ratably over the next five remaining months and we project generating on average about $8 million per month of free cash flow from our mining operations. We anticipate this to be sufficient to meet our remaining CapEx needs for 2025, which we project to be under $100 million. Importantly, this number does not include any near-term HPC and AI capital needs, which are planned to be funded by Macquarie. In closing, we believe our financial position provides a solid foundation to execute on the HPC and AI initiatives that Ben laid out this morning, and we look forward to keeping you updated as we continue to scale our U.S. operation. With that, I’ll turn the call back to the operator for questions-and-answers.

Operator, Operator

Thank you. Operator provided instructions. Our first question comes from the line of Mike Colonnese of H.C. Wainwright & Company. Please go ahead, Mike.

Mike Colonnese, Analyst

Hi. Good morning, guys, and thank you for taking my questions. First one for me is on the HPC, AI side. Jeff, you highlighted $100 million or so of CapEx that's not including any additional development of the HPC data centers. But if you could provide maybe some more specifics as to what the infrastructure development to continue to build out the 500 megawatts of total expected capacity would entail, that total number? And then, from an infrastructure development standpoint, what would need to happen thereafter in terms of client discussions for you guys to bring the site to the next stage?

Jeffrey Lucas, Chief Financial Officer

So, let me start off initially here by just commenting that we're actually working now in the final stages of our master development plan, what we're going to do with Panther Creek. So, we're actually going to have a lot more information forthcoming shortly here in terms of what those expenditures are going to be. But right now, we're still in the process of putting together the expectations there. Ben, maybe you want to add about some of the timelines and expectations that we have there?

Ben Gagnon, Chief Executive Officer and Director

Yeah, happy to and thanks for the question, Mike. So first thing is, as Jeff said, we're completing our master site plan, supposed to be done this quarter and that's one of the next key development milestones in order to access the full $300 million facility with Macquarie. Just give you a ballpark of what that $300 million can accomplish for us at that Panther Creek site. Effectively, we can build out all of the powered land infrastructure for the full 500 megawatts there at the Panther Creek location, or we can build out the initial substation and the first building of HPC for the power campus and we'd need further financing to finish out the remaining five or six buildings there.

Mike Colonnese, Analyst

Got it. That's very helpful, Ben. Appreciate that. And then, on the Bitcoin mining side, you guys have obviously improved the fleet efficiency quite a bit through the transformative fleet upgrade here to 19 joules per terahash. But curious, is there any further room to optimize or improve your overall power costs on the Bitcoin mining side?

Ben Gagnon, Chief Executive Officer and Director

So we have two different things there. I mean, first, we have about 6% of the miners still remaining to be installed. All of those miners that are waiting to be installed have efficiency higher than the 19 joules per terahash. There's going to be marginal improvement there on the energy efficiency side. And I think the big part of this here is that with the PJM facilities that we've secured and the active mining that we're doing throughout Pennsylvania, we now have a lot of energy trading opportunities in our portfolio. So for the first time, we have a lot more control and flexibility of our power prices. What this means is that we're not seeking to achieve a certain dollar per megawatt hour figure; what we're seeking to do is optimize free cash flow out of those mining operations on a daily basis. And so some days it might make sense to pay higher price and flex a little bit more hash rate. Some days it might make sense to run less and target a lower price. So with those PJM facilities, we have a lot more control over energy and that gives us a lot more downside protection and a lot more optimization on free cash flow. And that's our big way here of managing hash costs rolling forward.

Mike Colonnese, Analyst

Thanks, Ben.

Jeffrey Lucas, Chief Financial Officer

Thanks, Mike.

Operator, Operator

Thank you. Our next question comes from the line of Thomas Shinske of Cantor. Please go ahead. Thomas?

Thomas Shinske, Analyst

Hi, Ben and Jeff. This is Thomas on for Brett. Thank you for taking my question. I guess you just mentioned the master plan as part of the additional milestones for the additional $250 million secondary tranche from Macquarie. I guess what specific other development milestones must be met to access the additional $250 million?

Jeffrey Lucas, Chief Financial Officer

Yes. As you said, the milestones are pretty straightforward. The way it works here is that as we put together the master development plan, we've identified what the anticipated capital expenditures are going to be along the way. Once we have the approval from Macquarie for the master development plan, we can begin drawing upon that facility to take advantage of each of the various tranches as the capital requirements arise. So it's actually a pretty straightforward mechanism.

Thomas Shinske, Analyst

Awesome. Thanks. And then just one more, if I may. Ben, you mentioned as we enter the second half of 2025, you will look to start securing some critical substation infrastructure. I was just wondering if you've already put yourself in the queue for some of the long lead items here given the high demand for this infrastructure?

Ben Gagnon, Chief Executive Officer and Director

Thanks, Tom. We're still working on the master plans. We haven't placed deposits on any of the equipment yet. There are timelines here, so we should expect to have the first tranches of additional power probably in 12 to 18 months from today at the Panther Creek location and that includes our anticipation that we'd be putting orders down for substation equipment in the second half of this year.

Thomas Shinske, Analyst

Awesome. Thanks, guys.

Operator, Operator

Thank you. Our next question comes from the line of Mike Grondahl of Northland Capital Markets. Please go ahead, Mike.

Mike Grondahl, Analyst

Hey, guys. Thanks. It seems like you're off to a nice start with Panther Creek. How would you rank Scrubgrass, Sharon and Washington State? Do you guys have an internal ranking or will it be customer demand that drives what's next? Just kind of like your thoughts there.

Ben Gagnon, Chief Executive Officer and Director

Thanks, Mike. It's a great question. We've had a lot of initial conversations with potential customers and Panther Creek stands out as the best opportunity site that we have in our portfolio, that's due to a couple of reasons. One, it's got the most amount of immediate power over the next 12 to 18 months. Two, it has a very clear runway here for the next three years to scale up to several hundred megawatts. And three, its proximity to New York and Philadelphia is probably the best geographical location that we have in our portfolio for HPC, most in demand. As we look across our other sites, Washington, Scrubgrass, Sharon, the sites have different qualities and attractiveness to different kinds of customers. Scrubgrass is very similar in footprint and demand as Panther Creek, but Panther Creek is a little bit further along in the additional power connections to the grid that we're going to be able to use to scale up to several hundred megawatts; effectively almost all of the same characteristics as Panther Creek, just Panther Creek is a little further advanced in timeline. Sharon is an interesting one as well because it has 110 megawatts there that should be ready in the back half of next year, and that's a very attractive one as well. It's got the size and scale to facilitate a lot of interest from customers, but it doesn't have the same further expansion capacity that really makes Panther Creek and Scrubgrass such exciting prospects with that multi-year development ramp up to almost 500 megawatts. Washington is a very different kind of site. It's a much smaller site, so it'd be targeting a different kind of customer, but it also has probably the most cost effective and reliable power in our portfolio. For HPC in Washington, the expected price is about $26 or $27 per megawatt hour. So that's the lowest possible cost we have across our entire portfolio. It means that we'd have very healthy margins. We'd be going after a different kind of customer, probably something like a government agency, who'd be interested in taking the whole site and having the data sovereignty and the security also themselves. So the different sites have different attractive qualities for different potential customers. And clearly, Panther Creek is our flagship one that we're going to develop. But after we've completed our master site plans this quarter, we're going to be replicating similar efforts across all of those other sites in the U.S. so that we have a pipeline of sites to market to customers simultaneously.

Mike Grondahl, Analyst

Got it. That's helpful. And then, Slide 14 has CapEx needs for 2025 less than $100 million excluding HPC, that less than $100 million, what is it going for? Is that a little bit more infrastructure? Just help us where that's being spent?

Jeffrey Lucas, Chief Financial Officer

Sure. Let me start off and then Ben can add a little color if he wishes. First of all, we actually have that amount pegged at about $75 million at this point in time, of which roughly $5 million is logistics costs associated with our miners as we have further deployment at strategic locations. On top of that, the balance is primarily going to be spent on electrical infrastructure and to a degree on Scrubgrass and Panther locations as a sort of building out the electrical infrastructure. So roughly around $65 million for that overall. So that's kind of where we have it allocated at this point in time.

Mike Grondahl, Analyst

Got it. Okay. Hey, thanks for that color, Jeff.

Operator, Operator

Thank you. Our next question comes from the line of Brian Kinstlinger of Alliance Global Partners. Please go ahead, Brian.

Brian Kinstlinger, Analyst

Great. Thanks so much. Last quarter, you mentioned the hyperscalers would go through a quick assessment initially to see if the location is worth dedicating their time for a more thorough evaluation. At what point and what do they need to see to start that initial assessment?

Ben Gagnon, Chief Executive Officer and Director

So thanks for the question. We had initial conversations with potential customers even before we had closed on the Stronghold transaction. One of the key things for large scale potential customers like that is the certainty around their investments. It's really important for us to first secure the sites before those advance to the next level. The next big thing here for us is completing the master site plan because that includes timeline expectations, budget estimates, and Gantt charts for how construction builds out over time, what the phases of development are and expected power draws at what points in time. That is really the next key thing that we need. With that in place, it's a lot easier to have serious conversations with potential customers.

Brian Kinstlinger, Analyst

Great. And my follow-up, you've sold 60% to 70% of your Bitcoins mined over the last two quarters. With the financing in place for HPC and AI, and your limited CapEx for Bitcoin mining, how should we think about Bitcoin HODL versus planned sales going forward?

Jeffrey Lucas, Chief Financial Officer

For the sales going forward, we plan to use proceeds to address our operating requirements, which are roughly about the fixed costs that we face of about $6 million to $6.2 million a month. Above and beyond that, we're going to be looking to accumulate a HODL position. We are also looking to raise strategically and tactically for a Bitcoin One program, as a means of using derivatives and instruments like that to enhance the value and the quantity of Bitcoin that we hold.

Operator, Operator

Thank you. Our next question comes from Nick Giles of B. Riley Securities. Please go ahead, Nick.

Fedor Shabalin, Analyst

Thank you very much, operator, and good morning, everyone. This is Fedor Shabalin on behalf of Nick Giles. Ben and Jeff, my first one is about macro environment. Where are you now in the process of discussion with potential tenants for your HPC capacity and how would you frame up the interest for capacity in PJM specifically? Thank you.

Ben Gagnon, Chief Executive Officer and Director

Thanks for the question. We've had initial conversations with customers to gauge their interest, but what we really needed to do to take those conversations to the next stage is to have the certainty and the timelines for energization, power availability, construction schedule and associated budgets and costs. All of that is in the planning stages now in the master site plan development phase, which should be completed this quarter. Then we'll be ramping up efforts quite aggressively, not just at Panther Creek, but replicating the same process of doing a master site plan for each site and seeking site-specific financing partners so that we can put those sites into a pipeline to market to customers simultaneously.

Fedor Shabalin, Analyst

Thank you for this color. My next one is, you already finished the assessment of potential HPC sites. What would be the CapEx needs per 1 megawatt of growth capacity? And what are you looking for in terms of your prospective customers for this HPC capacity? Are you expecting kind of a CapEx split for development of HPC megawatts between you and potential tenants? Any color here would be helpful.

Ben Gagnon, Chief Executive Officer and Director

Good question. When you look at how we're driving shareholder value, the big unlock comes from long-term contracted revenues rather than the specific facets of development. There are a variety of business structures and financing models we can use to secure those contracted revenues, and there's a balancing act between CapEx associated with the model, the revenues that are possible per megawatt, and the margins on each unit of energy. Think of it as a spectrum: the lowest cost is powered land in the few hundreds of thousands of dollars per megawatt. A powered shell is probably in the low $2 million to $4 million per megawatt range. To move further up the stack to a completed building ready to put racks into, it's closer to $8 million to $10 million per megawatt. We think the sweet spot for Panther Creek will likely be the powered shell range. That seems to be where the balance of CapEx, revenue per megawatt and margin sits in most conversations. Different models will make sense for different sites and customers — some hyperscalers may want to take more control of the build and operations, while other customers or government agencies may prefer turnkey solutions. The primary next steps are completing master site plans and having financing in place to begin construction and conversion across all sites.

Fedor Shabalin, Analyst

Thank you, Ben. That's extremely helpful. And I promise the last one, it's a quick follow-up on Macquarie financing. So I know someone already asked it, but want to understand it a little bit better. $50 million is initial draw and $250 million is conditional. Again, could you clarify what milestones must be achieved to be able to draw the next tranches and what is the size of these tranches? Just want to understand the timeline here. Thank you.

Ben Gagnon, Chief Executive Officer and Director

There's two tranches: the initial $50 million and the $250 million for a total of $300 million. One key development milestone for the $250 million is the completion of the master site planning. That includes engineering schematics, designs, site maps, budgets and Gantt charts for multi-year construction. It's a comprehensive development plan we need in order to access the full $300 million under that facility.

Jeffrey Lucas, Chief Financial Officer

To add, when we specifically draw, that's as the capital needs arise. Once we have the master development plan in place, we can present specific project investments to Macquarie and draw on the facility as payments unfold and as the financing needs arise for each project tranche.

Fedor Shabalin, Analyst

Thank you, Ben and Jeff. Appreciate all the color. Best of luck.

Operator, Operator

Thank you. Our next question comes from the line of Martin Toner of ATB Capital Markets. Please go ahead, Martin.

Martin Toner, Analyst

Good morning, everyone. Thanks so much for taking my question. I believe the Macquarie financing is for two years. Just wondering where you think you'll be in terms of development at that time and/or slightly before that? And what options do you think there will be to refinance and increase financing from there?

Ben Gagnon, Chief Executive Officer and Director

So the facility, as I said, would be enough in our calculations right now to either do the first HPC data center building completely, and some infrastructure beyond that, or enough for us to do all of the powered land infrastructure for the entire site. As we continue to develop and make progress on the HPC site, we should expect that opportunities for raising further capital at increasingly lower costs will be available. As we progress through the project and get closer to contracted revenues and free cash flow on the site, we'll be seeking more financing at that time and expect lower cost of capital.

Martin Toner, Analyst

Super. Thank you. What conditions would motivate you to build out infrastructure for the entire site as opposed to going for that first dollar of rental revenue, which seems important for lower cost financing going forward?

Ben Gagnon, Chief Executive Officer and Director

Really what we're focusing on now is the infrastructure that will service any final direction — substation infrastructure, transmission lines, and civil works preparing the land for construction. Those steps need to be done regardless of the final approach. Ultimately, potential customer demand and timelines will determine priorities — whether we prioritize the first building or more extensive powered land work. It's highly likely the first building will be prioritized, but that's dependent on customer conversations that will accelerate in the second quarter.

Martin Toner, Analyst

Super. Congrats to the team on the progress. That's all for me.

Operator, Operator

Thank you. Our next question comes from the line of Bill Papanastasiou of KBW. Please go ahead, Bill.

Bill Papanastasiou, Analyst

Yeah. Good morning, and thanks for taking my questions. For the first one here, Ben, perhaps you can take a minute to speak to your outlook for Bitcoin over the medium term and how that could translate to the Bitcoin mining industry? A number of peers including Bitfarms have been cooling CapEx spend and dilutive financing recently. Perhaps you can share your thoughts on the prospect for attracting investor dollars here going forward? Thanks.

Ben Gagnon, Chief Executive Officer and Director

Hey, Bill. We actually have held the same outlook on Bitcoin mining economics since about mid-2023. Our view was to take advantage of low hardware prices and to get hardware in place for the 2025 bull run, which we've completed. It was always our view that 2025 would be a big opportunity for miners and that we would look to diversify into other revenue streams around 2026. As of mid-2025, we've achieved our objectives to position ourselves for the expected cycle. We provided a sensitivity table in our Q4 deck showing our expectations for Bitcoin price and network cash rate through 2026 and how that maps to implied hash prices and direct mining margins. We're confident that the Bitcoin mining infrastructure and hardware efficiency we have will generate healthy margins and free cash flow through 2026. Our current view is that the best opportunity to invest capital is electrical infrastructure specifically for HPC and AI, rather than further Bitcoin mining at this time.

Bill Papanastasiou, Analyst

Awesome. Always appreciate your color on Bitcoin mining, Ben. And now just taking a step back and running through our progress with the transition from some of your Bitcoin mining plans to bringing AI HPC capacity online and securing a long-term partner, what would you say are the biggest lessons learned over the past year or so? Thanks.

Ben Gagnon, Chief Executive Officer and Director

We started with expertise in energy infrastructure and development. Some new lessons were understanding the different levels of requirements and quality controls around HPC, which haven't been prevalent for us previously in Bitcoin mining. That's why we engaged advisors to accelerate our learning curve. One example: we have a site in Quebec next to a former industrial plant with a large coal chimney that raised concerns about structural integrity of adjacent property for HPC development. It's the kind of adjacent property risk that hadn't been a major factor for Bitcoin mining. With our advisors WWT and ASG, we've rapidly learned what's necessary, completed assessments of our U.S. sites, prioritized them, and developed a repeatable process for development and marketing to potential customers. We're in a strong place now in the development process.

Operator, Operator

Thank you. Our next question comes from the line of Brian Dobson of Clear Street LLC. Please go ahead, Brian.

Brian Dobson, Analyst

Yeah. Thanks very much for taking my question. As you speak to consultants and potential clients, do you think, just from a very high level, you could characterize the willingness to devote resources on the enterprises and hyperscalers for near-term data center expansion? There's been some hangar in the industry that this type of spending is slowing down. Do you think you could just address that?

Ben Gagnon, Chief Executive Officer and Director

The demand for HPC data centers is very strong. While there have been headlines around individual companies, those have not significantly impacted demand in our conversations. Most customers are moving at full speed. The investments and multi-year development timelines involved — often hundreds of millions to billions of dollars — mean customers can't be distracted by short-term headlines. The demand we're seeing matches the secular dynamics of compute growth: the bottleneck is power, not chips. The McKinsey report I cited outlines 156 gigawatts of expected demand for HPC versus roughly 15 to 20 gigawatts in the Bitcoin network today — almost a 10x increase. We do not see slowing demand or investment; we see ongoing, substantial demand for power and compute.

Operator, Operator

Thank you. Our next question comes from the line of Joe Flynn of Compass Point. Please go ahead, Joe.

Joe Flynn, Analyst

I wonder if you could go into more detail on the comment you made regarding Panther Creek being best suited for a powered shell approach. Just wondering how you came to that decision, and if you see opportunities maybe to bypass some of the due diligence process we've seen in the market for more fully built-out sites. And then in that regard, how would you think about is the goal to ultimately pre-lease the site, which then the Macquarie capital would become the equity component of the financing or how do you think about next steps in that regard? Thanks.

Ben Gagnon, Chief Executive Officer and Director

Our primary goal is unlocking shareholder value through long-term, high-margin contracted revenues. We focus on return on invested capital, which means balancing CapEx and revenue. Often the powered shell is the sweet spot: it requires less CapEx than a turnkey build but enables strong margins and faster conversion of megawatts. The decision on exact approach will be driven by customer demand and the ROIC considerations. Some customers may want more turnkey solutions; others may take more of the build themselves. We'll do what's best to maximize shareholder value given those dynamics and the customer conversations that continue to unfold.

Joe Flynn, Analyst

Great. Thanks.

Operator, Operator

I would now like to turn the conference back to Ben Gagnon for closing remarks. Sir?

Ben Gagnon, Chief Executive Officer and Director

Thank you, everyone, for joining today. Just like to reiterate our recent strategic moves to rapidly transform this company. We've quickly become a U.S.-focused energy and compute company and we have a strong underlying Bitcoin mining business with a lot of exciting potential to convert to HPC and AI. We will look forward to keeping you up to date on our progress, and thank you very much for attending the call.

Operator, Operator

This concludes today's conference call. Thank you for participating. You may now disconnect.