Cipher Mining Inc. (CIFR) Q2 2026 Earnings Report, Transcript and Summary

Ladies and gentlemen, thank you for standing by. Welcome to Cipher Digital's Second Quarter 2026 Business Update Conference Call. [Operator Instructions] Please be advised that today's conference is being recorded. I would like now to turn the conference over to Courtney Knight, Head of Investor Relations. Please go ahead.

Good morning, and thank you for joining us on this conference call to address Cipher Digital's business update for the second quarter of 2026. Joining me on the call today are Tyler Page, Chief Executive Officer; and Greg Mumford, Chief Financial Officer. Please note that our press release and presentation can be found on the Investor Relations section of the company's website, where this conference call will also be simultaneously webcast. Please also note that this conference call is the property of Cipher Digital, and any taping or other reproduction is expressly prohibited without prior consent. Before we start, I'd like to remind you that the following discussion as well as our press release and presentation contain forward-looking statements. These statements include, but are not limited to, Cipher's financial outlook, business plans and objectives and other future events and developments, including statements about the market potential of our business operations, potential competition and our goals and strategies. Forward-looking statements and risks in this conference call, including responses to your questions, are based on current expectations as of today, and Cipher assumes no obligation to update or revise them, whether as a result of new developments or otherwise, except as required by law. Additionally, the following discussion may contain non-GAAP financial measures. We may use non-GAAP measures to describe the way in which we manage and operate our business. We reconcile non-GAAP measures to the most directly comparable GAAP measures, and you are encouraged to examine those reconciliations, which are filed at the end of our earnings release issued earlier this morning. I will now turn the call over to our CEO, Tyler Page. Tyler?

Thanks, Courtney. Good morning, everyone, and thank you for joining us today. I'm Tyler Page, CEO of Cipher Digital, and I'm pleased to welcome you to our second quarter 2026 business update call. Execution has been exceptional across the business this quarter, and we continue to build in line with our transformational strategy. What has become increasingly clear over the past few months is how each step forward in the progress of our flywheel is now reinforcing the next. The leases we've signed are giving prospective tenants more confidence to come to the table. The financings we've completed are strengthening our standing with capital markets and validating the colocation strategy we've laid out. And the construction milestones we continue to hit on schedule or ahead of schedule are reinforcing the trust hyperscalers place in us as a partner for their next data center campuses. Put simply, this business is building on itself in a way that compounds and the results are starting to show up, giving us an even clearer path to scale much larger. For those newer to our story, let me frame quickly what Cipher Digital is today. We control the full value chain of developing and delivering turnkey data centers to hyperscalers, from land and power origination to engineering, construction and operations, which is what allows us to move at the speed and precision hyperscalers require. Those capabilities have translated into a stable and reliable business with longevity. Three data center campuses leased to some of the most sophisticated technology companies in the world, representing billions of dollars of contracted revenue locked in over the next decade plus. And beyond those initial campuses, our pipeline keeps expanding with approximately 4.4 gigawatts of expected future developments, giving us years of visibility into growth well beyond what's already under contract. Let's zoom out and look at the full scope of what we've built. Across our operating contracted and future pipeline developments, the portfolio now totals approximately 5.3 gigawatts of capacity spread across 11 sites. The overwhelming majority of that sits in our pipeline, representing substantial growth beyond existing contracts. The remainder reflects our already contracted HPC capacity as well as our legacy bitcoin mining capacity at Odessa. Texas remains the center of gravity for this portfolio, and that is deliberate. Early on, we made the call that Texas would become one of the most sought-after regions in the country for large-scale AI infrastructure, and that conviction has been proven out. This quarter, we added up to 1.1 gigawatts of potential new future capacity in Texas with a new 900-megawatt site called Apollo and a planned 200-megawatt expansion at our current Stingray site. We are positioned exceptionally well for multiple levers of future growth via continued site acquisition and existing site expansion, as well as the addition of behind-the-meter generation, which is particularly well suited to our sites and not yet included in this pipeline. Now let's turn to the cash flow profile behind the contracted leases on Slide 5. Our 3 executed data center campus leases are expected to generate approximately $793 million of average annualized net operating income from October 2026 through September 2036. This slide shows the future of our revenue model. Stable, visible and contracted net operating income from long-term agreements with investment-grade counterparties. As we move into the second half of 2026, we are turning these projected cash flows into reality as our first rental payments begin at Barber Lake and Black Pearl. And I'm pleased to say that reality is arriving even sooner than we planned at our Black Pearl site. Slide 6 highlights the key developments from the second quarter, which have built momentum and position Cipher well for the second half of 2026 and beyond. First and most significantly, I'm proud to announce the early delivery of data center capacity at Black Pearl, 2 months ahead of schedule. At the request of our tenant, we executed an amendment to the lease that accelerated the development time line of initial capacity. I'm pleased to report that we delivered that capacity and rent has commenced at the site. I want to spend a moment on why this matters beyond the headline. Any developer can sign a lease. Far fewer can deliver ahead of schedule when a tenant asks for a faster time line without cutting corners or sacrificing quality. What we have now proven is that Cipher has the operational depth across all of our teams to compress a delivery time line on demand and without compromise. We believe this early delivery will pay dividends in two ways. First, this strengthens our credibility with existing and prospective tenants, giving them even more confidence to work with us at other sites in our 4.4 gigawatt pipeline. Second, it's an early proof point on construction execution that will serve us well in future financings since we can now point to a demonstrated track record of timely delivery. We will continue to prove our differentiation in construction and execution as we deliver on the rest of our current projects and sign new leases in the future, each step forward, reinforcing the next. The second highlight is our landmark financing of the Stingray data center. We priced an $810 million bond offering at a 6% coupon, fully funding Stingray through substantial completion. Greg will provide more information on the financing in his remarks, but it is notable that this third project-level bond issuance priced tighter than our previous bonds despite spreads widening in the credit markets, reflecting continued confidence in our story. Next, I'm pleased to announce that we've acquired an option on a new site we're calling Apollo. As our 11th site, the name is a fitting nod to Apollo 11 and to the same reach-for-the-impossible spirit that put footprints on the moon. This site provides up to 900 megawatts by 2031 and is located within 25 miles of San Antonio, Texas. The site has been submitted as a studied load in Batch Zero under ERCOT's updated interconnection process and its flat buildable terrain and proximity to San Antonio make it well suited for large-scale data center development. Finally, we continue to invest in our team. This quarter, we welcomed Bill Blevins as our Head of Grid Strategies, who previously served as Director of Grid Coordination at ERCOT, overseeing large load interconnections, experience directly relevant to how we navigate interconnection across our pipeline. We also welcomed Mohammed Abouelella, who joins us from Google, where he was responsible for technical due diligence, design and delivery across more than 5 gigawatts of data center capacity globally. He spent 11 years designing Google's build-to-suit and colocation data center campuses, and we're thrilled to have his expertise as we continue to build and sign new leases. These are just 2 of many additions we've made to the team this quarter, and they reflect how we are building the organization needed to execute on gigawatts of HPC development in the years ahead. These milestones, an accelerated delivery, a tightly priced financing, a new site secured on attractive terms and a strengthening team is what compounding momentum looks like in practice. Now let's take a closer look at our current development portfolio. Starting with Black Pearl. As mentioned, we executed an amendment to our lease with the tenant to accelerate the development time line of initial capacity at the tenant's request. I'm proud to say our team delivered on that commitment. The first data center capacity at Black Pearl was delivered in August, 2 full months ahead of the original schedule and rent has commenced at the site. Beyond the first delivery, the rest of the site continues to progress well towards the same previously agreed deadlines and milestones. The remaining data halls in Phase 1 are moving through mechanical, electrical and plumbing fit-out, while Phase 2 is advancing in parallel with concrete foundations, structural steel and underground electrical work all underway. On the procurement side, we've secured approximately 96% of the equipment required across both Phase 1 and Phase 2, giving us strong visibility into completing the remainder of the site. Now moving to Barber Lake. We're pleased to share that Phase 1, comprising approximately 168 critical IT megawatts remains on track with rental payments expected to commence in October. Our tenant has commenced beneficial use of the facility, including partial occupancy of the building and deployment of network racks. We've also now secured 100% of the equipment required to complete the project, giving us strong visibility of our path to completion. What was steel in open ground a few months ago is now a data center campus advancing toward completion. We look forward to providing further updates on Barber Lake's progress over the coming months as we work to deliver the first phase of data center capacity in September. We next turn to Stingray, which continues to move through its early construction phases. Earthwork, grading and pad preparation are progressing on schedule and underground electrical work has commenced at the site. We expect to begin concrete foundations and steel erection in the third quarter. With the project now fully financed through substantial completion and approximately 75% of the equipment secured, we're well positioned to keep the construction progress moving efficiently toward our expected delivery in the first half of 2027. We look forward to providing updates on Stingray's progress as construction ramps over the coming quarters. Odessa, our last operating bitcoin mining site, performed well in the second quarter. Today, we are operating 207 megawatts of capacity, generating approximately 11.6 exahash per second of total hash rate at a fleet efficiency of approximately 17.2 joules per terahash. In the second quarter, we mined approximately 346 bitcoin at Odessa. Importantly, we don't anticipate additional capital investment in this part of the business as we continue prioritizing HPC, and we are encouraged by the level of interest we're seeing in conversion of Odessa to an HPC site. We are having early-stage discussions with multiple prospective tenants. And while it's too early to share specifics, we look forward to providing updates as these conversations progress. Let's now shift to an update on our development pipeline. Starting with Odessa. This 207-megawatt site is already energized and currently operating under a fixed price power purchase agreement with Vistra Luminant. We're encouraged by the HPC tenant interest here and are in early-stage discussions with multiple prospective tenants. The appeal is straightforward as the site is already energized and converting it into an HPC data center represents a meaningfully shorter time line to power than a typical greenfield development. Reveille and Ulysses are both fully interconnection approved and not part of ERCOT's batch process. We are engaged in HPC hosting lease discussions with a broad range of tenants at these sites, and we remain focused on securing the right deal for Cipher, not just the first deal available. Looking further out, Colchis, Mikeska and McLennan, totaling potentially 2 gigawatts of gross capacity remain on track through ERCOT's interconnection process. Given that ERCOT is expected to finalize the batch process decision soon, we are sharing updates based on information we know as of today. All 3 sites have necessary deposits funded, land secured and their requisite studies and executed FEAs were submitted to ERCOT on time. We have strong conviction that all 3 sites will be included in Batch Zero, and we look forward to updating the market. Let's now look at the full picture of what this portfolio represents today as well as the new additions from last quarter. On the operating and contracted side, we remain at 907 megawatts. We expect Reveille and Ulysses to add 270 gross megawatts in 2027. We expect Colchis, Mikeska and McLennan to add another 2 gigawatts in 2028 and 2029. And looking to 2030 and beyond, we expect energization at Milsing and our new site Apollo as well as expansions at our Barber Lake and Stingray sites to add up to an additional 2.1 gigawatts. Both Apollo and the Stingray expansion have been submitted as studied loads in Batch Zero. The 500-megawatt Barber Lake expansion is expected to be in Batch Zero and Milsing is expected to be in Batch 1 as studies are still being finalized. Cipher's total portfolio now spans approximately 5.3 gigawatts across 11 sites. We are one of the largest developers of hyperscale infrastructure in the country with a contracted revenue base measured in the billions, a pipeline measured in gigawatts and a team that has now proven quarter after quarter the ability to turn opportunity into reality. That's the platform we've built, and we believe it's a platform that will define the next chapter of AI infrastructure development in this country. With that, I'll turn the call over to our CFO, Greg Mumford, who will walk you through our financing activities, capital structure and financial results for the second quarter. Greg?

Thank you, Tyler, and good morning, everyone. Tyler outlined the exciting momentum across our development platform. I want to highlight the disciplined capital model supporting that growth and the progress we've made this quarter. As mentioned on prior calls, our strategy is to finance contracted projects at the project level, preserve flexibility at the parent and optimize the capital structure as assets stabilize. Barber Lake established that the projects could be financed. Black Pearl demonstrated its repeatability. Stingray, our third successful project financing and our lowest coupon to date demonstrated scalability and improving capital efficiency. That progression gives us increased confidence in our ability to finance contracted growth and fund investment in our development assets. Let's take a look at our current capital structure and liquidity position. In June, we successfully completed an $810 million project level senior secured notes offering, funding the development through substantial completion. The transaction funded approximately 98% of project costs and reimbursed Cipher for $56.7 million of previously funded project expenditures. Like our prior transactions, the 5-non-call-2 structure preserves flexibility to optimize the capital structure as the asset stabilizes. The financing was approximately 8x oversubscribed and priced at a 6% coupon, our lowest to date. We now have completed 3 project level financings that fully fund our contracted obligations through completion. Our notes are structured to amortize during the base lease terms, aligning debt service with the cash flow generated by the leases. At the corporate level, we have a 4-year committed revolving credit facility for $200 million with a $50 million accordion feature supporting working capital and LC issuance. We also have 2 unsecured convertible notes totaling $1.47 billion. As of June 30, 2026, aggregate principal amount of corporate and project debt outstanding was just over $6 billion with no cash borrowings on our revolver. Let's now turn to a review of our financial results for the second quarter of 2026. Revenue for the second quarter was $25 million, down from $35 million in Q1, reflecting the decommissioning of bitcoin mining at Black Pearl, in line with our transition toward contracted data center revenue. For the quarter, we reported a GAAP net loss of $268 million or $0.65 per diluted share compared to a GAAP net loss of $114 million or $0.28 per diluted share last quarter. The quarter-over-quarter increase in net loss was primarily driven by a $150.5 million noncash warrant remeasurement loss compared with a $43.6 million noncash gain in the prior quarter. Compensation and benefits rose $7.4 million sequentially, primarily reflecting higher stock-based compensation, associated employer payroll taxes triggered by equity vesting and continued investment in the team required to execute on our contracted portfolio and development assets. As Tyler mentioned, these hires were critical additions, bringing in expertise that will be instrumental as we continue to execute and scale as a best-in-class HPC developer. General and administrative expense increased primarily due to higher legal, insurance and other costs associated with the continued build-out of the platform. Moving below the operating line, we generated $36 million of interest income in the quarter, reflecting higher average cash balances following the Black Pearl and Stingray financings. Interest expense was $67 million, up from $59 million last quarter, reflecting a full quarter of interest on Black Pearl compute notes. Now let's turn to our balance sheet as of June 30, 2026. We closed the second quarter with total assets of $7.5 billion, up $3.2 billion or roughly 75% from $4.3 billion as of December 31. The increase is almost entirely a story of capital raised and capital deployed. First, let's focus on the financing proceeds. In February, we closed the Black Pearl senior secured notes and in June, we closed Stingray. Together, those transactions drove restricted cash up to a total increase of $1.7 billion. Net of cash deployed during the year, that is $1.4 billion in current restricted cash and $264 million in noncurrent debt service reserve funding. At quarter end, restricted project cash totaled approximately $3.7 billion, including approximately $3.2 billion reserved for construction and approximately $526 million for DSRA and interest during construction funding. Unrestricted cash was up an additional $204 million to $832 million, helped by $289 million of CapEx reimbursements embedded in the Black Pearl and Stingray financings. Our total unrestricted liquidity position stood at $870 million, comprised of $832 million of unrestricted cash and cash equivalents and $38 million of bitcoin. This excludes undrawn revolver availability. We remain well capitalized to execute on our near-term commitments and based on current forecast, we do not expect to require additional equity. Second, let's look at the build, where we are firmly in execution mode. Property and equipment rose $1.5 billion to $2.13 billion. Construction in progress grew $1.4 billion to $1.68 billion, reflecting the simultaneous construction ramp at Barber Lake, Black Pearl and Stingray. On the liability side, accounts payable grew to $289 million at quarter end from $40 million at year-end and accrued expenses and other current liabilities grew similarly to $357 million from $90 million. Both moves reflect the same story. The simultaneous construction ramp across Barber Lake, Black Pearl and Stingray, along with the timing of project billings, accruals and vendor payments. Put simply, these are the balance sheet dynamics you'd expect from a company in active construction mode. The company continues to make strong progress across the development platform, and we remain well positioned from a liquidity perspective to continue to execute and invest in future growth. Before we open the call for questions, I want to reiterate our commitment to disciplined execution, capital allocation and delivering long-term value for our shareholders. Putting the quarter in context, we now have a 4.4 gigawatt pipeline, a financing model that has proven itself 3x over and the liquidity to remain agile. We look forward to keeping you updated on our progress in the quarters ahead. Thank you for your continued support. Tyler and I would be pleased to take your questions.

[Operator Instructions] And our first question will come from Solomon Thompson with Morgan Stanley.

It's Stephen Byrd, actually. Congrats on a really constructive update. I wanted to -- can you hear me okay?

Yes, I hear you, Stephen.

Okay. Perfect. I wanted to first get your initial reaction. I know none of us have had that much time to absorb the letter that Governor Abbott issued yesterday. But I wondered if I could just get your initial reactions and implications for Cipher. Obviously, you have a bunch of assets kind of across the range from those that have no impact and therefore, might go up in value to those that are right in the middle of the process. I'm just curious any initial reactions you might have.

Yes, sure. So listen, like a lot of folks, we were a little bit surprised to get the Governor's letter yesterday. But I think we weren't surprised in the sense that the theme is very consistent with the batch process, and all the challenges that Texas is dealing with right now. So specifically, Texas is the most sought-after place for data centers right now because it's the best environment. It's got the best setup. And they have been a bit of a victim of their own success as they try to whittle down the queue and just sheer volume of development that is sought there. And so what I'd say is this is a sort of further reinforcement of the theme behind the batch process, which is the people of Texas, the legislature in Texas, the Governor of Texas, the citizens of Texas, need a way to determine who is serious and who is going to be a welcome addition to the grid, to the neighborhood, et cetera, and who is less serious and not going to be sort of fulfilling their obligations as a good neighbor and good grid citizen. And so the batch process is already underway. And so I think one thing that is unfortunate is that we're not going to get the answers this Friday that we were hoping for. But I think the governor's themes are really important ones, which highlighting things like auditing the attestations that have been submitted by everyone to make sure they're legit and ensuring people are behaving the right way by participating in things like the water surveys that he highlighted. I'll highlight for Cipher, we've done all those things. We've already submitted water surveys, and we stand behind every attestation we made in the process. So I think it's too early to tell the exact impact other than to say -- I think there's 3 takeaways that I think about and how it might impact Cipher. The first and most important one is whatever the finalized process and time line looks like here, Cipher is going to be at the front of it, okay? We have excellent people on our team monitoring these developments. We try to be very good neighbors, and we try to be in front of exactly the kind of issues the governor is highlighting. And we generally support any actions that will help separate the serious from the less serious because Cipher is serious. And so however the process gets finalized, our sites that we think are at the front of the line will remain there over time. We will do everything to ensure that happens. I think the other thing I'd highlight is that we already have client interest at the sites where we haven't even finished the Batch Zero process. And the demand environment is very strong and extending. So while we're not going to get the answer on Friday we wanted, whenever it comes, the demand curve is going out further at better terms, and our sites are going to be at the front of that process. So in that sense, disappointed, we won't have an answer Friday. Beyond that, I think we're still really well positioned in the process. I think the second takeaway is that the letter and the time line now implies that the value of any near-term megawatts outside of that process just went up. I have never seen a better environment for us in terms of how lease terms are evolving with higher rents, longer time periods, triple net structures, et cetera. And keep in mind, Cipher has 477 megawatts potentially available in 2027 outside of this process. And we are pretty involved in discussions with multiple tenants, potential tenants, I should say, at those sites and the terms logically should improve if there's an unknown time line to go through the batch process. So that's fantastic for our portfolio, to your point, Stephen. And then lastly, I'd say sites where we're going to bring our own generation, some people, you'd refer to it as behind the meter, but any sites where we're producing our own generation, anything we can do on that front is also now more valuable. And I'll highlight what I've said before that all of our sites have potential ingredients for that to be a massive success. We have some of our best people at Cipher working every day on the bring-your-own generation solutions. We have access to natural gas and very excited potential tenants for that. So stay tuned because I think that is also more valuable. And most of those structures envision setups where eventually, some of that generation is exportable to the grid. So I think it's in line with the challenges that Texas is trying to address and certainly what the governor's letter is trying to address. So long answer is it actually doesn't change anything about how bullish I am for Cipher. Short-term frustration is we're not going to hear on Friday.

That's super clear. Well thought out. Maybe just one follow-up on the last point you mentioned, Tyler, just on behind-the-meter generation. I did want to just get your sort of a temperature check on how important is that to customers in the sense of providing a site that's just much bigger than grid access alone. It has other advantages as well. So basically, how excited are you as a tool that your customers want you to use? And how meaningful could that be as we think about your growth?

I mean, it's potentially extraordinarily meaningful. And I say that because the raw ingredients are there. And what do I mean by the raw ingredients. We have the world's most highly rated companies very interested in getting -- I don't want to say as much as possible, but very large data center capacity. We have sites with, let's say, readily available access to extraordinarily large quantities of natural gas. And we are working very hard at solving all of the engineering challenges, supply chain challenges, financing challenges that come by pulling together sort of generation creation and data center creation. I think the potential is larger than the rest of our portfolio. But how that actually comes into being will be determined by a lot of things like supply chain and financing and engineering. And it's too early for me to give exact forecasts on size, but it's potentially enormous.

And our next question will come from Paul Golding with Macquarie.

Tyler and Greg, congrats on all the progress and fantastic execution. I wanted to start off with a question on Reveille and Ulysses. So 2027 target energization is still on track. How are conversations going with prospective tenants for those sites? We heard the detail you gave on Odessa as a potential conversion site. But given the near-term energization targets for the other 2 sites that are not yet interconnect -- energized, let's say, since they are interconnect approved, wondering how those conversations are going and the extent to which you might be able to get ahead of energization for construction if you were to do a deal? And then I have a follow-up on Apollo.

Sure. So I'd say the demand environment has never been stronger. We have multiple interested parties in both sites. We mentioned that given this increasing level of demand and the backdrop of, frankly, a new scarcity element from some of the developments with the approval process in Texas, we want to make sure we strike the best possible deal. Best possible deal includes both great structures, great terms, long-term leases, triple net structures, high rental rates. Also, we're monitoring what's going on in the credit markets and the read-through to the credit quality of the counterparties. So thoughtful structuring around how to get the best lowest risk returns for Cipher is sort of top of mind there, but the demand is robust. So it's just a matter of getting to a place with picking our dance partner. I expect all those available megawatts will end up leased.

Great. And then on Apollo, it's great to see that it's been included in Batch Zero. Could you give us some context on how that option came about and how you were able to execute on seemingly the inclusion in the Batch Zero process while also negotiating that option?

Sure. I think this is a real testament to the strength of our deal team. We -- they have been extraordinarily busy as things have progressed in Texas because we have seen, as we mentioned, I think, on a previous call, all kinds of opportunities get created by the batch process. So again, zooming out, the thinking behind the process is determining who is a serious developer. And one of the proxies being used for that is the ability to post necessary collateral deposits and do them in a timely fashion. And so the team has been extraordinarily busy. We passed on a ton of deals that we did a lot of diligence on. This particular deal met our diligence criteria and sort of fit a scenario that was very unique that only Cipher could execute on. This is a development site where there was a need for a pretty large deposit to be posted in a very short time frame. Our team was able to do quick due diligence, assess the likelihood of the site to end up in Batch Zero, and we're able to then structure a very, very, very cheap price on the site because we had to come up with a decent chunk of money to post the deposit on a short time frame. So we structured it as an option. What's interesting about the developments with the letter yesterday and so forth is that we do not have money at risk. If for some reason, there was an odd twist and turn in the process in Texas, we get all our money back, the deposit we posted, et cetera. We don't have to exercise the option. On the other hand, if things progress like we hope they will reasonably quickly, we think this is an absolutely fabulous site at a fantastic price. So I'll highlight that it depends a little bit on what shakes out in the overall progress of the process in Texas. But I think the bigger point, and this is a point I try to make to our stockholders a lot, we are not like a closed-end fund with a handful of development sites. We have a true growth equity story. Our team will continue to originate best-in-class deals where we control downside risk and produce an extraordinary ROI. And this deal is indicative of that. I'm excited about this site. I'm hoping that we get clear direction in Texas in the near term. If for some reason we don't, a big takeaway for shareholders should be that Cipher can structure very favorable deals in any environment. And this validates our standing in the industry as a partner that people want to work with, someone that can move quickly, be sophisticated, has access to capital, et cetera. So excited about Apollo, looking forward to the direction things take.

And the next question is going to come from Bill Papanastasiou with Chardan Capital Markets.

Congrats on the strong execution at Black Pearl and being able to adapt at your tenant's request. Also great to be back on the call. Tyler, I was just hoping to get an overview of the demand environment that you're seeing given you have boots on the ground and how that demand funnel is shaping up recently. How is the team weighing who to select as the next tenant to partner with? Do you see yourselves as a launch pad for one key tenant? Or is diversification top of mind? How are you navigating that?

Let me give my answer, Bill, and then maybe I'll call on Greg to give some color on sort of financing markets because those are dynamic and moving around and a key part of the answer to how we think about it. But first of all, what I'd say is to highlight something I mentioned in the prepared remarks on the call that there is an element of a flywheel taking shape to our business, where success begets more success, which begets better financing, which begets more deals, et cetera. And so on the one hand, we are working with the very best tenants in the world at our sites. We are in a world awash in stories about delays, we are delivering early. We stand out. I can't highlight that enough that I think there's been questions around our time lines, how aggressive they are, et cetera. And even against those aggressive time lines, we've now delivered early for one of the most demanding tenants in the world. Obviously, that requires a lot of close work with that tenant over time, their engineering team, our engineering team. And it's very natural that if we have a lot of success and a great working relationship, it is our hope and goal to do more repeat business with those tenants. And I think that's likely to happen in the future. That said, we are known in the industry. Everyone chatters. We have seen an extraordinary amount of increased demand going forward. You've seen the read-throughs on the increasing CapEx budgets for the hyperscalers. That is absolutely consistent with what we are seeing behind the scenes. So let me be unequivocal. This is the best demand environment we've ever seen. Terms continue to improve for developers. We have to juxtapose those improving terms against how our model works, which is financing needs at the project level. I would say the demand environment we're seeing now very much validates our choice to focus on colocation and building full turnkey data centers for the very best tenants in the world because those terms are getting even better than we were already getting. And as there are questions around financing as more and more debt comes to market around this sector, the colocation model is very, very solid. I mean, Greg, maybe I'll call on you to add a little bit of color just from what you're seeing coming ahead in the financing markets and maybe how we're thinking about positioning ourselves.

Yes, for sure. Thanks, Tyler. And Bill, great to have you back on the call. It's good to hear from you. So what I would add to that is certainly, we are watching what's happening in the credit markets. It is a fundamental part of our business, being able to go out and finance these transactions. So we're seeing the treasuries moving up. We're seeing widening spreads. And certainly, there's a lot of supply as it relates to AI infrastructure debt coming to market. And as you have more supply, you expect eventually to have a more discerning investor thinking about where they're putting their capital. So I mean, I think this really plays to our strengths and plays to our model. First, I would highlight that our committed funding that we already have in place for our existing projects is fixed rate, and it's kind of medium term, it's 5-year fixed rate debt. So the current rates and current market doesn't affect that. As we think about the future, the way that we really sit down and look at a deal and you talked about do we want to be a launch platform for one tenant or do we want to think about some diversification, we evaluate each deal in the context of that deal and in the context of the market that we're in at the time. So what I mean by that is we'll look at, first and foremost, the quality of the counterparty, that's critical. But then it also comes down to things like the development parameters. So what's the design complexity that they want for us to go out and build and how does that affect the supply chain? Does our team think that we can execute on the schedule that we put forward. And then there's the risk profile of the actual contract. So we look at each contract and not every contract is the same. There's different outs, there's different milestones, things of that nature. So we want each project to stand on its own when we finance it. That's why to date, we have done every project level financing on a nonrecourse basis, and all of them are structured to amortize during the base term of the lease. We need each project to stand on its own, and it needs to generate attractive risk-adjusted returns. And so we will continue to review tenants under that context when we look at a deal.

Appreciate that color, Tyler and Greg. And then just high level, what's the strategy today on looking at out-of-state opportunities? You picked up that Ulysses site fairly recently. Is the power team seeing anything meaningful in Ohio or any other states outside of Texas?

I think some of the challenges are the evolving requirements in different locations. And so Ulysses was a structure that is in PJM, but is grandfathered in before their new deposit requirements around collateral. So we look at everything that comes across the transom. I think there's no question that most of the team's time has been focused on Texas over the last quarter as people get ready for the batch process and there's folks that may have had a great site, but not access to capital to post the necessary deposits. We look at sites all over. I'd say the huge focus, though, remains Texas. And I'd say the other sites we look at generally are evenly distributed across a handful of locations across the South and the West.

And the next question will come from Richard Choe with JPMorgan.

I just wanted to get a little bit more detail on how you were able to deliver 2 months early. And is that something that can be done if the customer really wants it and seems like to do that sort of quick delivery is pretty hard to do in general. But just wanted to get more detail there. And then if you can talk a little bit about the -- any changes to the equipment procurement process.

Yes. So I mean, I think high level, let me give kudos to our excellent in-house construction, engineering, procurement, operations team. They are largely ex-hyperscaler. They have worked at those shops. They work very well with the types of tenants we've got. And so I'd say the real secret here, there's kind of 2 things. One, you've got to have a team that can see around corners and work well with those types of tenants. We certainly have that. I'll also highlight that at that particular site, we have a setup where we have a cap on expenses. So there is an element of saying like could we have more burst labor here to do some things faster? And obviously, we've got a willingness to do that if we find a way to pay for it, typically, that doesn't fall on us. So beyond that, there's things like leveraging supply chain partnerships, the fact that we do in-house procurement and then have that team to refine the design every day really allows us to work together with tenants. So is it repeatable? Yes, in the sense that the ingredients are there with a team and a willing tenant that wants to work towards accelerated schedules, but there is a cost to that. So we'd have to negotiate that. So -- and I think the other thing is, too, there's an element where I don't think you're necessarily pulling forward an entire data center build. But all of these builds become somewhat iterative as you go through the process. You may have change orders. People want to change the scheduling of when things are delivered. It may not be the entire data center. So I do think opportunistically, in general, it's just a testament to the team here that we were able to do that against the backdrop of delays.

The next question is going to come from Jon Petersen with Jefferies.

Maybe continuing on there. So if you are ahead of schedule at Black Pearl, I guess, Phase 1, does that put Phase 2 ahead of schedule too? How should we think about that?

No, I think it's fair to generalize that the rest of the delivery deadlines are currently -- they currently remain the same at Black Pearl. This was getting a piece of it ready early to get the tenant on site and working on elements of it. But the rest of the delivery schedule, we expect will be online with the same time line.

Okay. And then on the ERCOT batch approval process, I know a lot of it is kind of up in the air, but can you maybe frame a base case expectation? What's kind of a best case scenario for the news that we might get, I guess, pending delays from the governor's letter. But -- and yes, I guess, like kind of like what's the base case and what's the upside for you guys?

So look, that's really hard to predict because over time, that process is important enough to all the interested constituents in Texas that it has generally gotten extended. And so look, we were expecting this Friday to be the deadline when we would get the results, and that would open a path to progress more quickly at the sites that we think have the great positioning to be in baseload in Batch Zero. It's very hard to say it's pretty fresh, right? The letter came out yesterday afternoon. We've certainly spoken to all of our contacts in Texas to try to get a read other than it's very serious, clearly, because of the attention it's getting. It's hard to make exact predictions on the timing of what that means. Again, I kind of repeat myself, but I think no matter what that timing is, I expect Cipher to be very buttoned up on top of it and standing at the front of the line. But at this point, it's just a little too fresh to have a serious prediction of the timing. I just don't know if it's like the governor has requested an audit of those waiting in the interconnection queue. Depending on how long that audit takes or if we get more color on how significant and substantial that will be, that may give an indication of timing, but it's too hard for us to predict the exact timing. I do think what will happen though is that, that audit logically would decrease the amount of load in Batch Zero, certainly the baseload piece. And so I think it's going to make it even more important that you are a development team that is very buttoned up and has dotted your I's and crossed your T's and paid your deposits. And so again, that should strengthen our relative position. I just can't predict the exact timing because it's all still pretty fresh.

Okay. And if I could sneak in one on the balance sheet. You guys said in your prepared remarks that your business plan didn't call for new equity. Was that kind of a shorter-term time line? Or are you generating enough revenue now that you have revenue coming online from these projects where you can organically generate that revenue from recurring cash flow to finance future deals?

Yes. It's Greg. I'll jump in and take that one. So I think that's really a function of a point in time and a function of what happens in our development platform. So if I look back at the last quarter, in Q2, we spent about $400 million in CapEx, about $50 million of that was on the development pipeline. So whether that's acquiring land, building out electrical infrastructure, certain grade sites, basically just getting things ready to a point where it would be equally as attractive to any tenant. Those CapEx expenditures can scale up and down, but that's not including the equity downstroke you would actually need if you were to go out and sign a very large deal, whether that's multiple hundreds of megawatts up to a gigawatt. Depending on what happens on the development pipeline and how leases materialize, there may be a situation down the road where we do need equity. But looking at where we're at today, $870 million of liquidity on the balance sheet. We have a revolver that we're not drawing, we have no cash draws on the revolver to date. We feel very good about where we're at. We feel like we could handle the sites that are right in front of us without needing to tap the equity market.

And the next question will come from Mike Colonnese with H.C. Wainwright.

Congrats on all the progress here. First one for me is on the development pipeline. You guys have obviously built out a very robust one. And just curious, based on your conversations with current prospective tenants for this future capacity, what would you say are some of the key development milestones that need to be realized before they're comfortable with executing a lease at a given campus. And it would be helpful to know what portion of the sites in your pipeline has reached a stage where you think they would be lease execution-ready in current form.

So yes, thanks, Mike. So I'd say, listen, generally, historically, the dividing line has been whether or not you have that final interconnection approval because otherwise, you are waiting for that key input. So if you look at what we've got available today, again, it's Odessa, Reveille, Ulysses, 477 megawatts. All of those are past whatever milestones they would need to get a lease. So the milestones remaining are, do we like the terms and the tenant and we work through the legal documents. I expect that they will all be signed. Beyond that, there's a sort of a newer development, and I think it speaks to the demand environment. Historically, again, tenants had not been interested pre-interconnection approval in talking about a site. We have had tenants interested in the sites awaiting Batch Zero. I think particularly Colchis is pretty juicy as a 1 gigawatt site. We've already had lease discussions on that site with interested parties. We'll have to see how things develop around the batch process. It's not inconceivable we come up with some sort of condition precedent structure or something that would allow us to move forward given that that's a 1 gigawatt site. That said, that's all new territory and speaks to just how serious the demand environment is. It kind of remains to be seen. But so the short punchy answer is the 3 sites with 477 megawatts available are all very leasable. It just depends on when we get to a deal we like. And then beyond that, there's some upside as we await the process in Texas. And that's outside of any bring-your-own generation sites, which, again, has some deal complexity. But in general, the driving desire for the tenant behind those structures is to get a more accelerated time line. So if we pull that together, the types of discussions we're having are, are there ways to get my first megawatts online in calendar year 2027. In order to do that, we would have to make quick progress on a lease, just given building time lines. But again, it kind of remains to be seen the overall scope of that opportunity, but it's very, very real.

Very helpful color, Tyler. I appreciate that. And just a follow-up for me. It sounds like you're getting some really strong interest in Odessa. Just curious what the potential data center delivery time line look like if you were to sign a lease and convert Odessa over to an HPC campus for bitcoin mining.

Yes. So much of that is going to depend on when exactly we would come to a full agreement. But I think, obviously, we have a site that's already constructed there that is a containerized data center for bitcoin mining. So should be relatively easy to decommission that site. And if we were to come to terms soon, certainly, we would be hoping to get our first megawatts online in calendar year 2027. But that will depend on getting a lease done in the coming couple of months so that we don't get too far, or I should say, too close in terms of building time line to having something ready. So very end of 2027 would be kind of the target there.

And the next question comes from Chris Brendler with Rosenblatt Securities.

Congrats on all the progress here, amazing execution. I wanted to ask on the equipment procurement details you're giving us by site. These percentages are pretty impressive and also pretty impressive disclosure. Can you just give us more of a qualitative update on that process? I imagine it's getting more and more difficult given all the construction demand out there for data centers. How -- can you give us some qualitative comments on the equipment procurement side?

Yes. Let me start by saying that a big focus in some of our hiring over the last quarter has been deepening our bench focused on the procurement activities here in-house. It is something different with our setup versus most of our competitors. Most of our competitors outsource procurement. We still manage that supply chain in-house. So Chris, one of the reasons you get more transparency from us is that we actually know where we stand with procurement at the sites. We're not sort of subject to the vagaries of the market and hoping to hear something positive from our outsourced provider. We've added a lot of depth to that team because that's such a key activity for us to manage. And I have to give general color that I think there's a high level of demand for everything. There is no question that from time to time, we'll see something harder to acquire, and we'll see decent amounts of inflation and cost. But in general, when we build a construction time line, we work backwards from the longest lead time items, and those are generally established, and we know where we stand before we agree to a time line with a tenant. Again, as a broad generalization then, as we're going through a building process, if a particular piece of equipment becomes highly in demand, harder to procure, we can kind of manage around those challenges.

Makes sense. And just a follow-up there would be, as we progress here and hopefully sign additional HPC contracts for these additional sites, do you expect material increase in your power CapEx per critical IT? Are we still around $10 million a megawatt? Or is that heading north from here?

No, it's really going to depend on the tenant's build specs. So I've highlighted this sometimes in some conversations with investors that we iterate so much on the demands of the particular tenant and their understanding of a particular site that that's really what drives that. We are also awesome at procurement and managing our relationships, but it's not like we're necessarily so much better at getting a cheap price on transformers. It's that we are able to have a discussion with tenants about like, hey, given the historical reliability at this site, do you really need backup gens in the quantity that your standard basis of design would call for? And maybe the answer is no, and then maybe that drives a lower price. I mean -- but what I can generalize about is that we definitely see inflation on labor and equipment across the board. And so as we build budgets for the next build, I would expect our budgeting to go up, and that probably translates into a higher cost per megawatt, at least to compensate for inflation. But any particular site may be at the lower end or higher end of the cost curve. And that's of course contemplated when we put together a proposal for a lease.

Right. You usually get those costs back.

And the last question is going to come from John Todaro with Needham.

Michael Chen on for John Todaro. Yes, going off of that procurement, historically, we've seen you guys quote CapEx in the $9 million to $11 million per megawatt range. And I know a lot of the equipment has already been secured, but we've seen pricing pressure and supply constraints increase across the sector. So curious, as you guys are looking at future leases and seeing where equipment still needs to be procured, are you seeing that cost creep higher? And how kind of confident are you in still hitting that band going forward?

Yes. I mean, look, I think as it regards to our current builds, we put together a budget that has contingencies for stuff like that. And I would say we're running at the contingency level given that inflation. Going forward, we're building in even more because, as I just said, we are seeing inflation across the board for labor and for equipment. So I do expect that those numbers will creep higher. But again, at any particular build spec, I think it's hard to generalize because it's driven by the demands of the tenant. The other thing I'd mention, too, is keep in mind that some of our sites, we address this by putting a cap on the CapEx that we have exposure to. So sort of this is not really -- of course, we want to manage things as efficiently as possible, but we pass this risk to the tenant in some structures as well.

Got it. Understood. And on Odessa, if I'm not mistaken, the PPA there was set to expire. So is there a time line that you guys need to get a lease done in? Or was the PPA resigned there?

No. I mean that PPA runs through the end of July of next year, so 2027. Look, the attractiveness of that site is the ability to reconfigure it as an HPC site before the end of calendar year 2027. But any lease of that site for HPC will require a renegotiated PPA with Vistra Luminant, which is co-located with the site and providing power to the site. So that will all be hand-in-hand with the necessary lease discussions there.

And I will now turn the call back over to Tyler for closing remarks.

Thank you, everyone, for dialing in to our business update call. The future is extremely bright at Cipher, and we look forward to talking to you soon. Cheers.

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