Soluna Holdings, Inc. (SLNH) Q2 2026 Earnings Report, Transcript and Summary

Hello, everyone. Thank you for joining us, and welcome to the Soluna Second Quarter 2026 Earnings Conference Call. [Operator Instructions] I will now hand the conference over to Mike Tu, Vice President of Finance. Mike, please go ahead.

Good afternoon, and thank you for joining Soluna's Second Quarter 2026 Earnings Call. Our earnings release and the accompanying presentation are available in the Investor Relations section of solunacomputing.com, and this call is being webcast with the presentation. With me today are John Belizaire, Chief Executive Officer; Ryan Carver, Chief Development Officer; and Michael Picchi, Chief Financial Officer. Before management begins their formal remarks, we would like to remind everyone that some statements we're making today may be considered forward-looking statements under securities laws and involve a number of risks and uncertainties. As a result, we caution you that there are a number of factors, many of which are beyond our control, which could cause actual results and events to differ materially from those described in the forward-looking statements. For more detailed risks, uncertainties, and assumptions relating to our forward-looking statements, please see the disclosures in our earnings release and public filings made with the Securities and Exchange Commission. We disclaim any obligation or undertaking to update forward-looking statements to reflect circumstances or events that occur after the date the forward-looking statements are made, except as required by law. We will also discuss non-GAAP financial metrics and encourage you to read our disclosures and the reconciliation tables to applicable GAAP measures in our earnings release carefully as you consider these metrics. We refer you to our filings with the Securities and Exchange Commission for detailed disclosures and descriptions of our business, as well as uncertainties and other variable circumstances, including, but not limited to, risks and uncertainties identified under the caption Risk Factors in our annual report on Form 10-K for the year ended December 31, 2025, and our other SEC filings. One note before we begin, effective this quarter, we changed the presentation of pass-through electricity costs from a net basis to a gross basis. That change increases both reported revenue and reported cost of revenue by the same amount and thus has no effect on gross profit, operating loss or net loss. Every effective slide is footnoted, and Mike will walk through the mechanics. With that, I'll turn it over to John.

Thanks, Mike Tu. Hello, and welcome to Soluna's second quarter results and business update. This is our first quarterly earnings webcast, and we're glad to have you all here. Let me start with the thesis because everything else today sits underneath it. Power is the primary constraint in the AI era -- not chips, not capital -- power and how quickly you can reach it. Soluna has secured long-term behind-the-meter access to power at scale, and we convert that access into contracted data center flows and data center cash flows by building directly on the site of renewable generation with stranded power by bypassing long grid queues. Four numbers frame where we stand today. Revenue grew 145% year-over-year, our fifth straight quarter of sequential growth. We have 192 megawatts of capacity under management, a proven cash-generating operating base, and we expect that to increase this summer to 206 megawatts. We have over 650 megawatts across 2 AI campuses in development at Kati 2 and Dorothy 3, both on-track to secure leases and construction. And our renewable power pipeline grew 47% this year to over 6.3 gigawatts, the scarce input for AI sourced all behind the meter. Now let me spend a moment on the second and third of those because this is the quarter they connected. We like to say power is the asset and compute follows, and that's the Soluna Way. In the second quarter, we took it all the way through at Project Dorothy. On April 1, we closed the acquisition of the 150-megawatt Briscoe Wind Farm. On April 15, we acquired Spring Lane Capital's interest in Dorothy 1A. And on May 19, we acquired Navitas' interest in Dorothy 1B. Soluna now owns the generation and the compute across all 50 megawatts of Project Dorothy 1. Everyone in this industry is racing to secure power right now, signing PPAs, taking queue positions, negotiating with utilities for capacity that shows up in 2029. We acquired the wind farm. So when a hyperscaler or neocloud customer performing diligence on one of our sites asked 2 questions and decide everything. How fast can you get me power and what is my energy cost? We answer both with an asset we own outright. Now that model produces 5 revenue streams and the mix is shifting deliberately. Hosting for Bitcoin miners is our largest business today. We build and operate the data center, creditworthy mining companies bring their machines, and that is the driver of our revenue. Proprietary Bitcoin mining is our own fleet. We sell daily and hold no treasury. Grid ancillary services pay us to be flexible, interruptible load, and we are compensated for acting as behind-the-meter flexible load for the grid operator. And Soluna Wind is our new addition this quarter, on-site wind generation at Briscoe with ERCOT substation and grid interconnection. And of course, AI high-performance computing, colocation and hosting for companies that need AI-ready capacity is where this company is going. That is the segment Ryan was hired to build. Everything we are doing this year sits on the 4 priorities. First, develop AI. Advance Kati 2 and Dorothy 3 to shovel-ready and tenant-ready and build a pipeline of AI-ready campuses designed for rapid deployment. Next is optimizing our projects. We are focused on energizing and ramping Kati 1 and driving profitability across the operating fleet through uptime, operational efficiency and disciplined cost management. Next is capital formation. We intend to fund pipeline growth and AI development through project level financing and strategic capital partnerships while maintaining balance sheet flexibility. And finally, we are focused on growing our pipeline. As I said earlier, we reached 6.3 gigawatts this quarter with more than 300 megawatts of the growth coming from expanded term sheets at 4 sites already in our portfolio. I want to say something about capital formation because it is a question I get most. We raised $159.4 million in the second quarter and an additional $23.6 million on our ATM program since quarter end, issuing about 18.8 million shares. And here is what that accomplished. 100% ownership to Project Dorothy 1, a 150-megawatt wind farm, the Kati 2 joint venture and land for Dorothy 3 and a clean capital structure. We retired the Series B entirely this quarter and paid out the accumulated dividends. Every dollar of that went into assets that are now on the balance sheet today. Going forward, the large AI builds are designed to be funded predominantly with project level debt collateralized by the data center and underwritten against contracted tenant cash flows. Mike Picchi will take you through that structure. So let's get into the quarter. On the business side, 4 things. We were added to the Russell 3000 and Russell 2000 indices in the latest reconstitution and new sell-side research coverage has been initiated on the company in recent weeks. Together with more formal quarterly communications you are seeing today, these reflect a deliberate effort to elevate Soluna's visibility and accessibility to institutional investors. We closed the $53 million acquisition of Briscoe Wind Farm on April 1, as I mentioned, and it's our first direct ownership of a renewable generation asset. We consolidated 100% of Project Dorothy 1A and 1B, strengthening the path toward Dorothy 3, our new AI campus. On the project side, our teams delivered across all 4 campuses. Kati 2 reached a definitive joint venture with Metrobloks, completed design development and signed a tenant letter of intent. Dorothy 3 secured a definitive land purchase agreement and advanced utility coordination. Kati 1 completed substantial construction, and Dorothy 1A and Sophie held capacity through a heavy summer curtailment window. Ryan will take you through the AI project highlights in detail shortly. Now I'd like to talk about the pipeline, which is our core asset. As of August 1, the total pipeline is approximately 6.3 gigawatts. That's up from 4.3 gigawatts earlier this year, and I want you to see how it is structured. 192 megawatts is operating, energized and generating revenue today. 14 megawatts is under construction, the final phase of Kati 1. We expect this will take us to 206 megawatts operating by the end of the summer. Approximately 1.6 gigawatts is in planning and development where PPA negotiations, ERCOT planning, AI feasibility work, and land acquisition activities are underway. And approximately 4.5 gigawatts is in assessment with our power partners. One more piece of the model. Our behind-the-meter structure keeps our data centers flexible. We can draw power from the renewable plant and from the grid, and we can provide ancillary services back to the grid. That flexibility is what gives us rapid time to interconnection. And we will cluster. We plan to use multiple generation assets in proximity to a single data center site, which is how a footprint that would otherwise support a fraction of the capacity becomes a 300-megawatt campus. We are using that approach at Kati 2 and Dorothy 3, and you will see it in our other sites. Before I turn to the road map, I want to address the recent announcements coming out of Texas. On August 3, Governor Abbott directed the PUCT and ERCOT to audit every data center in ERCOT's interconnection queue before approving new projects. That is a response to roughly 474 gigawatts of pending requests, about 90% of which are data centers. That mandate targets new studied loads in the interconnection queue. I want to share 4 points on where we, Soluna sit. First, our direct exposure is limited. The audit targets new studied load, roughly 146 megawatts of our capacity in Texas is already energized and both Dorothy 3 and Kati 2 build off adjacent energized interconnections. Second, our model fits what the state is screening for: no costly transmission upgrades, flexible interruptible load, new wind and solar, minimal water by design. We fund our own electrical infrastructure. And third, we see this as a tailwind for operators with live capacity. With the queue frozen for the audit and ERCOT's August 7 dispatch delayed, energized capacity gains value. And fourth, we welcome the review. We support a rigorous and consistent review process, and we are cooperating fully with the PUCT and ERCOT. So I want to close with what's on tap for the balance of the year. We're focused on completing Kati 1 with the final 14 megawatts. We plan to move from design development to construction documents at Kati 2. We're advancing the Dorothy 3 development and begin marketing to potential tenants. New Bitcoin hosting announcements we expect to take place at Kati 1. And we are also looking at new power purchase agreements underway with Rosa, Hedy, Ellen, Annie and new projects. Those are the milestones, and we will report against them next quarter. All right. Before I hand it over, everything you've heard so far, the wind farm, the buyouts, the capital, exists for one reason: converting our power position into contracted AI megawatts, which brings me to the newest member of our leadership team. On July 16, we appointed Ryan Carver as our Chief Development Officer. Ryan joins us from Microsoft, where he was most recently serving as Senior Director of AI Construction and Site Development, leading a construction P&L in the tens of billions of dollars across the company's AI data center development. There are very few people who have delivered hyperscale AI campuses end-to-end, and Ryan is certainly one of them. His mandate here is deliberately broad from site selection and development to engineering and construction and operations, all in one organization reporting to me. Ryan, welcome to Soluna and take it away.

Thanks, John. I'm really happy to be here at Soluna. Good afternoon, everybody. I'm 4 weeks in, so I'll keep the biography short and spend the time on the projects. As John mentioned, I spent more than 10 years at Microsoft, most recently as Senior Director of AI Construction and Site Development. Practically, that meant taking AI campuses from a piece of land to world-class operating data centers: power procurement, permitting, design, construction, commissioning, and handover to operations. The most notable program from my portfolio is called Fairwater in Mount Pleasant, Wisconsin. My background is building large complicated infrastructure on schedule and on budget. Why I came to Soluna? For the last few years, I've watched this industry run into the same wall. Chips you can buy, power you have to wait for and buildings that can be built. Most of the answers I saw amounted to getting in line, the queue position, the utility conversation, capacity that shows up many years later than what could be achieved. Soluna's answer is the most compelling one I've seen, put the data center behind the meter, colocating facilities with sustainable power generation sites that are already built, underutilized and already spilling energy. It's a shorter, traceably more sustainable path to the exact same megawatt. My remit covers the AI campuses we're preparing to develop. But first, I'm going to take you through the operating campuses Soluna has today. Kati 1 is our 83-megawatt campus in Willacy County, and it's nearly complete. K1A, as we call it, the Galaxy Digital portion, is 48 megawatts and held steady through the summer 4CP constraint window while meeting its ERS program requirements. K1B is the build-out. Phase 1 is 12 megawatts of Cormint containers and Phase 2, 9 megawatts of Soluna-designed data centers. Both are energized and operating currently. Phase 3, which is the final 14 megawatts, is in construction and running ahead of schedule. Kati 1 delivered its first positive gross profit this year. Site moves from capital consumption to cash generation exactly once, and this is the quarter Kati 1 did it. And moving into the Briscoe Wind Farm that John had mentioned previously. Briscoe is a 150-megawatt wind farm in West Texas. We acquired it April 1 for $53 million. It's Soluna's first direct ownership of a generation asset, and it's now integrated into our operations. We inherited a maintenance backlog with the acquisition, and we made a very deliberate decision to fix it immediately. This work was completed in Q3. Beyond the turbines, we completed our Q2 renewable energy credit sale and did substantial community work with county officials and local landowners, which matters for everything we intend to build in that footprint. Moving into Kati 2. Kati 2 sits directly across the street from Kati 1. At full build, it's expected to be over 350 megawatts. Phase 1 is 100-plus megawatts of critical IT capacity. Phase 2 will add another 250, a lot moved this quarter. We signed a definitive joint venture with Metrobloks on June 3. Soluna holds all Class A interests and serves as manager. On design, John told you on the Q1 presentation that Phase 1 was at 30% schematic. We're now nearly complete in design development. For anyone who does not live in this vocabulary, schematic designs to concept, design development is where you commit to the electrical and mechanical topology; and then on to construction documents, which is what you hand a contractor to build from. We also brought our general contractor on board during the quarter, so the people who will build the site are in the room while we finish designing it. On procurement, we signed the commitments with key electrical equipment suppliers for certain long-lead items. On every AI project I have built, those are the long poles. You have to place those orders early or you risk delays. On power and site infrastructure, engineering is underway to expand the Las Majadas substation by an additional 100 megawatts in support of future phases with those upgrades expected early next year. And we executed a gas pipeline access agreement to improve resiliency. For future phases, an additional 150-acre parcels under agreement and the purchase is nearing execution. We've also begun conversations with county officials on a potential tax abatement. On the tenant, we signed a letter of intent, and commercial terms and lease negotiations are currently underway while we finalize the design. We'll update the market when we have a definitive agreement to announce. In the meantime, the clear signal I can give you is what we're doing while we negotiate. Design is complete through development. The contractor is on board. Long-lead orders are being placed. Substation engineering is running in parallel. That is the work a lease requires, and we're doing it now. Now let's talk about our second AI project, Dorothy 3. Dorothy 3 is slated to deliver 300-plus megawatts of AI capacity on land adjacent to our existing Dorothy campus in West Texas. On land, we have 397 acres under contract to support the initial build-out. On design, we've begun preliminary master planning and mobilized design teams. We've also initiated long-lead equipment procurement activities, launched environmental, water, survey and fiber studies as well as advanced ERCOT integration work. We're also looking at on-site gas options. We currently have over 1.6 gigawatts of AI data center capacity in development. Two things about this quarter's expansion are worth your attention. The first is where the growth came from. 4 sites in our existing portfolio got larger. Project Rosa went from 187 to 242 megawatts. Project Hedy went from 120 to 198. Project Ellen went from 100 to 145 and Project Fei doubled and went from 120 to 240. That's an additional 300 megawatts added at sites where we already have done the work. The land is identified, the power agreements are signed, and we have exclusivity. Rosa, Fei, Hedy, and Ellen are now designated for AI workloads. We also allocated Project Grace, our 2-megawatt technical validation effort with the Siemens PTI team to Dorothy 3 capacity. The question we get from most investors: the wind doesn't blow all the time. So how does the data center behind a wind farm serve a tenant who needs power 24 hours a day, every day for 15 years? The answer is augmenting the wind farms with additional power generation resources that act as one integrated system. First, the renewable plant itself, sustainable energy that would otherwise be curtailed and wasted at power costs around $40 per megawatt hour. Second, the grid. Our behind-the-meter design lets the campus draw from the plant or from the grid, whichever the moment requires. That flexibility is what gets us to power in months instead of years. Third, firming. At Kati 2, we've executed an access agreement with a natural gas pipeline operator, and engineering on lateral to the site is underway; on-site generation that takes the campus to the availability an AI tenant contracts for. It is a genuinely interesting moment to be doing this work at the point where AI demand meets renewable energy that would otherwise be wasted. I'm very glad to be here. With that, I'll hand it over to Mike Picchi for the financials. Mike?

Thanks, Ryan, and welcome aboard. I'm going to spend most of my time on the balance sheet and capital. Let me start there. Liquidity strengthened materially over the quarter. We ended Q2 with $113 million in cash available for project development and operations, and our current ratio improved to 2.1x from 1.7x at Q1. Working capital is a positive $69.2 million. On the asset side, current assets grew 48% to $134.6 million, and net property, plant and equipment rose $58.3 million to $137.8 million. That's the Briscoe Wind assets and the Kati 1 construction coming on to the balance sheet. Total assets grew 54% to $293.5 million. On the liability side, total liabilities rose modestly to $81.8 million from $76.1 million. Within that, there's a reclassification worth explaining. Current liabilities increased 26% to $65.4 million, while long-term debt fell 81% to $3.0 million. That's not new borrowing. It's the reclassification of our generate debt from long term to current as we intend to repay that in the near term. In fact, earlier this week, we prepaid just over half of the loan outstanding. Total debt across the business at June 30 was $33.1 million. The short version, we ended the quarter with materially more liquidity, materially more owned infrastructure and a simpler capital structure than we started with. Turning to the P&L. Revenue was $15.1 million, up 145% year-over-year, a fifth straight quarter of sequential growth as sites and customers ramped. One accounting note that affects this line. Effective this quarter, we present pass-through electricity costs on a gross basis rather than net in both revenue and cost of data hosting. That's a change in presentation applied prospectively. It adds approximately $4.4 million to each line for the second quarter with no effect on gross profit, operating loss or net loss and prior quarters are not revised. So there are 2 ways to read the top line and both are correct. As reported, revenue grew 145%. Excluding the presentation change, revenue grew 73% year-over-year and 13% sequentially. The footnote is on every affected slide. The growth drivers include hosting revenue as Dorothy 2 reached full quarter operation compared to the prior year. Kati 1 contributed, Dorothy 1A ramped Blockware and Canaan and Dorothy 1B began hosting. Briscoe added $366,000 of wind revenue, net of the intercompany elimination. Offsetting that, proprietary mining declined $1.1 million or 40% on a 34% decline in hash price from roughly $51 to $34 and the deliberate conversion of Dorothy 1B capacity from mining to hosting. Gross profit was $766,000, which compressed 35% year-over-year from $1.2 million. Two things drove that, and I want to be specific because the revenue line moved the other way. New site costs came online ahead of full revenue contribution. Kati 1 is energizing in phases and the cost of running the site arrives before the site is full. And Briscoe's repair work ramped up. The wind farm carried roughly $1.5 million of turbine repairs and maintenance in its first quarter under our ownership and posted a gross loss of $787,000. Ryan walked you through why we chose to front-load that work. Underneath the consolidated number, the operating fleet performed. Data hosting delivered $1.9 million of segment gross profit, the largest contributor. Demand response contributed at effectively full margin. Proprietary mining posted a loss on hash price compression and Kati 1 turned its first-positive gross profit, the point at which a site moves from consuming capital to generating it. I'd also note that cost of revenue includes site level depreciation. When you consider EBITDA at the segment level, Q2 2026 saw 50% growth compared to Q2 2025. Net loss was $22.6 million in the second quarter compared with $7.8 million in Q2 2025. The increase is driven by non-cash and financing items that I'll describe shortly, partially offset by site level operating improvements. Adjusted EBITDA was a loss of $1.6 million, which improved 25% sequentially from a $2.1 million loss in Q1 and roughly flat year-over-year. Stepping back further, that's a 76% improvement from the $6.4 million loss in Q3 of 2025. The trend line is the point. Below the line, there are three items you'll see in the reconciliation of net loss to adjusted EBITDA. First, noncash stock-based compensation was $9.4 million, reflecting overlapping 2025 and 2026 equity awards amortizing on schedules set at the grant date. Second, interest expense of $3.2 million was up from $1.2 million a year ago. And third, there was a $4.2 million loss on debt extinguishment, $2 million related to the Generate Tranche B modification that partially funded Briscoe and $2 million related to the early Yorkville loan payoff. General and administrative expenses increased $9.8 million year-over-year, of which $7.5 million is the noncash stock compensation increase. Salaries and benefits added $1.1 million and professional and legal fees added $720,000, driven by the Briscoe transaction, hosting agreement negotiations, project financing and technical accounting support. Full reconciliations are in the appendix and the earnings release. Now to capital, which is the part I focus on and get the most questions. We raised $159 million in the second quarter. $113.5 million of that came from the ATM program, $24.5 million of debt financing, $18.9 million under the standby equity purchase agreement, which is now fully utilized and $2.5 million from warrant exercises. The debt was $12.5 million drawn on Generate's Tranche C to complete the Briscoe transaction and $12 million from the Yorkville note we drew and fully repaid within the quarter on June 12. And we deployed $159 million. $51.4 million net went to Briscoe Wind Farm, $25.3 million went to buy out our joint venture partners at Dorothy 1A and 1B. There was $17 million in debt repayment, $13 million of CapEx and equipment deposits and $7.4 million of working capital and other. Ultimately, $45.3 million of cash went to the balance sheet. That last number matters, and it's there for a reason I'll come to on the next slide. We also simplified the capital structure. All 62,500 shares of Series B preferred stock converted into 6,510,416 common shares at the adjusted $0.96 conversion price. We paid $2.1 million of accumulated dividends, and we filed a certificate of withdrawal on June 23. No Series B preferred stock remains outstanding. Subsequent to the quarter end, we issued approximately 18.8 million additional shares under the ATM program for net proceeds of $23.6 million. Now let me close on how we intend to fund what Ryan described. Winning an AI contract requires investment before the contract exists: fiber studies, environmental and geotechnical work, long-lead equipment, deposits, site design, joint venture formation, securing power. A customer performing diligence on a site is evaluating work that has already been done. You cannot wait to be awarded a contract and then begin. Our policy for the large build-out is project-level debt. For a 100-megawatt build-out, which we estimate at $1.2 billion to $1.3 billion, we would target 70% to 80% project level debt, likely high-yield notes sized against the contracted tenant cash flows under the lease, with the remaining 20% to 30% from Soluna contributions or third-party project equity. We would expect that debt and equity formation to occur in the 8- to 12-week period following announcement of a signed lease. Having capital already available to begin construction in that window is another use of cash balances we have built through the first 7 months of 2026. Bitcoin and AI are capitalized separately at the project level, which gives each access to capital appropriate to its risk profile. This is how we underwrite an AI lease. At the lease rates currently being struck in the market for critical IT capacity on long-duration triple net terms, a 100-megawatt 15-year contract would generate sufficient revenue and net operating income to service and retire the project level debt and deliver a strong multiple on the equity invested. The illustrative case on this slide shows roughly $180 million of stabilized annual net operating income against an estimated $1.2 billion to $1.3 billion build-out based on the assumptions disclosed here. When we do reach an agreement, the results may differ materially. I offer it to explain why we are investing ahead of a contract. It's not a forecast and consider that this is just the first 100 megawatts of our AI pipeline with expansion possibilities at each campus and additional projects to layer on top. In summary, Q2 was a quarter of asset consolidation. We took full ownership of Dorothy 1. We acquired the generation next to it. We formed the Kati 2 joint venture. We retired the Series B preferred stock and ended with the strongest liquidity position in the company's history. The operating fleet improved, adjusted EBITDA improved sequentially and the capital we deployed went into assets on the balance sheet. With that, I'll hand it back to John.

Thanks, Ryan and Mike. Let me close on the shape of the quarter. We grew the pipeline where we already hold power, expanding term sheets at Rosa, Hedy, Ellen, and Fei by more than 300 megawatts. We completed vertical integration at Dorothy 1. We acquired Briscoe Wind Farm on April 1, then bought out our partners at Dorothy 1A and 1B, allowing us to achieve vertical integration at the D1 site. We advanced both AI campuses, Kati 2 through design development and into construction documentation with a signed letter of intent for a prospective tenant and Dorothy 3 through land, fiber and utility studies, pipeline, power, projects and now customers. That's the sequence this business runs on. And in the second quarter, we moved on every one of them. All of it sits on clean energy that would otherwise be curtailed or wasted. That's what we mean by renewable computing. It's our mission to make renewable energy a global superpower using computing as a catalyst. Thank you for your time today. Operator, we'll take questions now.

[Operator Instructions] Your first question comes from the line of Michael Colonnese with H.C. Wainwright & Co.

Congrats on all the great progress across the platform from an execution and financing standpoint. First one for me, you announced last month that you signed an LOI with a potential tenant for Kati 2. Just curious if you could share more information on how those negotiations are progressing and what needs to take place before a lease can be executed at the campus and what investors could expect from a timing standpoint?

I'll start. Thanks, Michael, for joining the call. We get that question a lot, timing of the lease. As you know, these leases are very complex activities, really detailed negotiations need to take place, and that's all underway. So while we can't speak to timing, what we can say is that we're now executing like an infrastructure -- AI infrastructure company. We're building the execution teams, developing our sites, designing the data centers, ordering long lead equipment, as you've heard, starting preconstruction activities, doing community development and gearing up capital formation as we negotiate the commercial terms with at least one potential customer. So I'd say the answer to timing is stay tuned, and I'd say that we're making good progress in the process over here.

And a number of questions we've received from investors is around power availability to energize Soluna's near-term AI data center developments, particularly Kati 2 and Dorothy 3. And I appreciate the comments you shared there, John, around the location adjacent to existing energized facilities that are powering Bitcoin mining data centers. But how should investors think about current energized capacity that could theoretically power Kati 2, Dorothy 3 and then really the incremental megawatts needed to say, let's power the first initial phase for each of those developments, especially as you navigate some of the audit requirements that are going on in ERCOT. I know it's a mouthful, but it's come up quite a bit in our investor conversation.

Yes. No, we -- I can imagine that question would come our way. First of all, as I said on the call, we are very supportive of the audit process. Governor Abbott is looking at things the right way. Data centers should be positive additions to the community, to the grid and should be thinking about how they're going to affect those 2 elements. And so what we can say is that for the past 8 years, Soluna's design has been focused on that very thing, making a positive difference in the community, developing data centers that integrate and service the grid. And the fact that they're behind the meter gives us an advantage because we already integrate with generation assets. And that is exciting to us because the governor is sort of implying that that should be the way, and it's already the Soluna way. And so to your question, Dorothy 3 will source its power from the grid and the Briscoe Wind Farm. As I said, we now own that asset, and so we can deliver that energy quite easily to the Dorothy 3 project. And the Kati platform, the Kati campus that we're building out is drawing power from the Las Majadas wind farm. Also already energized as we had commented across the street at the Bitcoin side, allowing us to stay within this concept of energized loads being less of the focus and more of the focus on study loads. And so we'll be able to source energy from Las Majadas. We have worked with them to increase the amount of energy that we can draw from that power plant. And that's why we're doing additional upgrades to support that new power envelope, if you will, to support the campus expansion. The other thing that I mentioned in my portion of the conversation and Ryan made mention to it in the Kati 2 update, we are also investing in bringing firming energy to the site by drawing from local gas lines that are close to Kati 2, and that will allow us to look at the development of additional generation back there to provide additional prime power to the site. So both campuses have access energy. Both campuses are already energized as part of the previous investments we've made in grid integration. And so we feel we're positioned very well as part of the audit. Everyone is going to get an audit. We welcome it. We already know how we will answer those questions, and I think they will be well received by ERCOT and the governor's office.

Your next question comes from the line of Michael Donovan with Compass Point.

And also congrats on the progress. So you've previously discussed hyperscaler and neocloud interest at Kati 2. Can you help us characterize the prospective tenant now under LOI? And then at Dorothy 3, how would you characterize the quality and stage of inbound interest you're seeing?

Mike, do you want to take that? Just to share the love.

Thank you, John. Well, with regard to qualifying the Kati 2 interest, we have not specified whether it's a hyperscaler or neocloud that were under letter of intent. And so we'll leave it at that for today. And then with regard to Dorothy 3, I would say the inbound level of interest is very keen. In fact, I would say it pulled forward our development efforts there as we work through this year. We have been very focused on Kati 2, and we've talked a lot about Dorothy 3 here in the last quarter. Those 2 projects in aggregate, 650 megawatts, but now you see already us lining up the projects behind that, Project Fei and Rosa and the like. So the Dorothy 3 customer interest has accelerated our development of the site. And so we're very excited about that. Helping having the Briscoe Wind Farm there is a significant asset. It's going to make that a really wonderful campus for us.

Yes. And I would add that the types of customers we're seeing and the types of interest fills the gamut in terms of hyperscale and neocloud.

Appreciate that. And Mike, you mentioned pulling forward develop that Dorothy 3. So just for clarification, are you formally marketing Dorothy 3? Or are you still on track for beginning in fall? And if it's beginning of fall, what remaining milestones need to be completed before you launch the formal tenant process?

So I'll start there. What we've been doing is following a very specific process we've laid out that says we have to put a master plan together, have begun any kind of interconnection updates and amendments that we need to do to the interconnection process for AI and land, long lead, all the things that Ryan mentioned before we market. But given the level of demand that's happening in the marketplace right now, that doesn't keep people from calling us. And so we have been receiving inbound interest, and we're generally saying here's sort of some high-level things about the site, and we'll begin formal marketing in the fall. So I'd begin to continue thinking about that time line, but I also want to share that our phone is ringing.

Understand. Ryan, congrats on joining Soluna. You touched upon long lead items. How -- generally, how are you thinking about equipment availability and broader supply chain risk as you move toward construction?

With respect to equipment availability, we already have a lot of key long-lead equipment, electrical equipment on order right now. And that's always something that's a risk in the industry. I mean, every building have ever built, especially recently, equipment is always in question. So being prepared, preordering equipment as much in advance as possible as we can as an organization and looking at our campus master plans and looking to quantify how much equipment we need and when we're going to need it, so we can be prepared for it. And when it comes to water, right now, we're looking at leveraging closed-loop systems for the most part. We're not using direct evaporative right now. So water on site will not be an issue. We do have a couple of other -- we do have some wells that are drilled, and we've identified aquifers that can serve the admin and other common areas right now.

Great. And then one more, if I may. So obviously, this will be really driven by customer interest and needs. I was hoping we could get a bit more granular design and build-out of data centers. So once a lease is executed, how should we think about the phasing, sizing of individual data center halls and time line from lease signing to initial RFS?

I mean, the sizing of data halls, it's going to really -- that may differ depending on the client ask, right? And I don't want to get into any proprietary details at the moment. We're not prepared to speak to that naturally. What was the second part of your question, I'm sorry?

Yes. So thinking about time lines, getting to initial RFS.

So right now, we're looking at for Kati 2, it's going to be 15 months, roughly 15 months after contract signing.

Your next question comes from the line of Chris Brendler with Rosenblatt Securities.

On all this progress. A bit newer to the story. So forgive me if this is somewhat of a bigger picture question that may not be appropriate for a conference call, but I'm sort of struck by all the opportunities and all the execution you've had this quarter and then all the opportunities you have ahead. How do you think about allocating capital towards the growth opportunities that you're faced with? And is there any sort of idea of potentially partnering or growing the footprint through some sort of combination given your relative size and the amount of your pipeline. I'd love to hear like how you're thinking about that dynamic.

Thanks for the question. I think we do get that question a lot. We're punching above our weight, if you will. And our key asset, our pipeline continues to grow, how will we convert all of that asset base into spinning revenue given our size and access to capital. The way we think about it is the way we've always thought about it. We look at essentially stair stepping the growth of the business, proving one project at a time that we can execute. We'll get access to capital for that initial project. And then as we execute and start driving revenue, it opens more doors to capital as we are able to execute. We are always looking at both financial partners and execution partners that can help us to scale. That's why in our AI structures, we do explore joint ventures that can help accelerate us, give us access to expertise, capital and other assets that we may not have. And so when we look at the almost 1.6 gigawatts of AI projects that we have underway right now, we are very focused on the first 2 because that's going to be the proof point that we are, in fact, an AI infrastructure platform. We are building out a team to allow us to execute around those projects. And once we have done that, we will have access to more assets that can help us grow and attack the rest of the pipeline. And so I guess we think of it as being incremental. And through that approach, we believe that will help us to grow and be successful in our strategy.

That's fantastic. My follow-up question was on Texas. And I thought your answer on the reasons why you're well positioned given the current situation with ERCOT is pretty strong and reasons for optimism that you kind of hit the top of the queue there. Does the -- like in your experience, does the renewable generation that you bring sort of help escalate your requests? And is that something that you sort of like before this all happened, were you expecting to be 0? If you get any color on that front? I would just think that given your power profile and bring renewables when there's a lot of concern about making sure we don't impact consumers here and bringing power and renewable power is a great way to do that. So I just love to hear any thoughts on the intricacies of that process at ERCOT since you have some experience there, Ryan?

Yes. So that's as great question, Chris. I'll certainly take that. The way to think about it is we have built an incredible amount of process expertise and understanding of the ERCOT market. It's one of the things we do very well as a company. And early in the life of this entire large load process, we were one of the few companies that sat with the grid operator and provided information and data about these new types of data centers and technology. In this case, it was Bitcoin. And that's the insight that they needed to understand the modeling and what would make things successful by integrating these assets and also how to approach the large load process. We are among -- we were one of many, but we were one of the first to sort of open the kimono, if you will. What we like about this audit process, it's a continuation of that approach to transparency. We have, as a company, a core focus on our technology and our projects really being of service to the grid, as I mentioned. We deploy our data centers with renewable energy to consume wasted and otherwise unused power, which allows the grid to absorb more of that type of power. If you go to every one of our sites, we are a positive addition to the community. We don't use water. We create jobs. We bring our own power, which is behind-the-meter integration. And with the AI approach that we're doing, we're going to do more of that. And if you zoom in on what happens to power prices and whatnot in the regions where we're located, because of the structures and the way power flows and the revenues that flow to those communities, power costs actually go down. And so -- when I take a step back and look at the Soluna model versus my peers and other loads joining the queue, we're probably a blueprint for how this should be built going forward. And so we feel pretty positive about the fact that because our sites are already energized, we will be included in the batch process going forward. And we were already submitted as batch load. So it's kind of hard for a grid operator to say, you have power, you're using the power. Tomorrow, you no longer have that power. That's not the goal of this audit. It's really just to make sure that these projects are genuine and will be additive to the grid. Now we can't make any promises. This process has just launched. I think tomorrow is the day that they're going to sort of announce more detail on how it's going to work. But what we expect is we'll receive RFIs that essentially ask us about what are you guys doing back there? How does this affect the community? What's your background? How are you going to build these projects, time lines, budget, all that kind of stuff? And we're prepared to provide that information because we think it's the right thing to do.

Your next question comes from the line of John Roy with Water Tower Research.

Excellent. Great quarter. I wanted to maybe take a step back a little bit and look at it a little bit differently. If you look over the next 18 months, John, what do you see as the single biggest risk to achieving what you guys have laid out?

I'll start the answer, and I want to turn it over to the rest of my colleagues to provide their perspective. I always love the risk question because we ask ourselves the same question inside the company, what are our biggest risks and how are we going to mitigate it? We are transforming the company into an AI infrastructure company. We weren't one before. So in that transformation, we look at what are the biggest risks and challenges. And I think the #1 risk and challenges is execution capability. Do we have all of the tools, the architecture, the expertise, the talent the partnerships, the capital to be successful over the next 18 months. And when I zoom in on that, I think the biggest risk earlier this year was do we have the people, the expertise that can derisk our execution and allow us to perform at the highest possible levels. You know how you solve that? You go hire a Ryan Carver, you hire the best of the best and ask him to go build a team. And that's what we've been focusing on over the last quarter is expanding our team with expertise that we now need to go execute on this incredible asset that we have. I'll pass the baton to Ryan Carver to get his perspective, and I'm sure Mike Picchi has some thoughts.

Yes. Thanks, John. I would tend to agree with you that that would be the biggest risk, but you appropriately articulated it's something that we're building right now. We're building the team. I do have a very strong background, a lot of experience with hyperscale development at the highest level at Microsoft, and I'm very humbled to have that experience and be able to bring it over here to Soluna. So as John mentioned, we're building a team as far as risk is concerned. Someone already touched on long-lead equipment risk. That is something we can mitigate though, by getting in front of it, as I previously noted. I would say, continuing to build the team. That's our biggest risk, and we're mitigating it now.

Thanks, Ryan. I think in terms of the challenge in the data center industry, 2 big challenges, access to power and access to capital. And I was so attracted to Soluna because of the neat power position. And I just love that tie-in to that we get -- and we're bringing load to power. We're bringing load to the generation assets that already exist. We don't get in a line for the utility queue or the grid operator down the road. We're tying into these wind farms and solar farms that already exist. But the second part of that is the risk item I think about raising the capital. And I think at the project level, attack -- plan of attack and raising project-level debt at that roughly 80% loan-to-cost ratio is viable and has been proven multiple times by -- in multiple transactions over the last 5 years is a viable path. And then in terms of on the Soluna side, for the other 20%, coming up with the equity or equity partners or joint venture partners that could help fund the equity piece of it. And so that's what we work on and trying to have a portfolio approach to how we fund these projects that will allow us to execute on all 1,600 megawatts that are in line of sight right now and active. Thanks, John.

Great. Can I get a follow-up real quick to Ryan? I really wanted to understand, Ryan, what really holds you to Soluna and what really excites you about things going forward?

Soluna has a really interesting value proposition to me. When it comes to co-locating the data center directly adjacent to sustainable power assets, I think it was very, very interesting to me. And as Mike alluded to and John alluded to previously, we already have the power generation assets in some conditions and conditions, we own the entire life cycle from power generation to the data centers that we're going to be developing here very shortly. So there is a unique value prop for, again, co-locating data centers with renewable power generation that would significantly reduce the time. I'd love to build very, very fast. Safety is always the #1 priority, right? I think that goes without saying when you're in the construction industry but we can go even faster because of the fact that we own our own assets, and we can bring the data center right there. And there are a number of different wind farms where we already have pre-existing exclusivity agreements, PPAs are already negotiated. There are a lot of different advantages that we have where we have a lot of assets that we can take advantage of very, very quickly that we're not going to be blocked from when it comes to getting into an interconnection queue. So rapid deployment, sustainably traceable power consumption. Out there in the data center industry, there's a lot of different companies that are offsetting their environmental impact. There's a lot of people are going to sustainable carbon zero, net zero by 2030 and things of that nature. We have sustainably -- we have traceable, sustainable power that we can prove how much we're consuming, faster time to market, and we own the assets. So those are all very appealing things to me.

We have reached the end of the Q&A session. This concludes today's call. Thank you for attending. You may now disconnect.