Title

Presentation

Operator: Hello, and welcome to the Bit Digital First Quarter 2025 Earnings Conference Call. Good morning, good afternoon and good evening, depending on where you are joining us from. Thank you for being here. [Operator Instructions] Also, as a reminder, today’s conference is being recorded. I’ll now hand it over to your host, Cameron Schnier, Head of Investor Relations at Bit Digital. Cameron, the floor is yours.

Cameron Schnier: Thank you. Good morning, and welcome to the Bit Digital first quarter 2025 earnings call. Joining us on the call today are Sam Tabar, Chief Executive Officer; and Erke Huang, Chief Financial Officer. Before we begin, I would like to remind all participants that some of the statements we will be making today are forward-looking. These matters involve risks and uncertainties that could cause our results to differ materially from those projected in these statements. I, therefore, refer you to yesterday’s 10-K filing and our other SEC filings. Our comments today may also include non-GAAP financial measures. Additional details and reconciliation to the most directly comparable GAAP financial measures can be found in our 10-K filing, which is on our website. [Operator Instructions] With that covered, I will turn the call over to Sam to discuss our performance. Sam?

Sam Tabar: Thank you, Cam. Ladies and gentlemen, thank you for joining us on the call today. Today, I’ll walk through our first quarter results and provide key updates across the different business units at Bit Digital. Let’s start with the mining business. Our overall top line results were dragged down by our Mining segment, where first quarter 2025 revenue decreased 64% year-over-year and 26% sequentially. This contrasts greatly with our HPC business lines, which demonstrated solid growth. Mining results were affected by the 2024 halving event and by our fleet redeployment program as we exited Coinmint facilities at the end of 2024. These factors contributed to an 80% year-over-year decline in production to 83 Bitcoins for the quarter.

Despite the lower production, our mining operations remained gross margin positive. In fact, mining margins expanded approximately 500 basis points sequentially to 21%, reflecting improvements in fleet efficiency and cost structure. Our active hash rate stood at approximately 1.5 exahash by the end of March 2025 and fleet efficiency was approximately 24.5 joules per terahash. We had a shipment of previously ordered S21 miners from Southeast Asia that we paused amid tariff uncertainty as the prescribed import duties would have significantly increased payback periods. But we have since begun taking delivery of those units and expect to return to approximately 2.5 exahash with fleet efficiency in the low 20s during June. Mining represented just 31% of our total revenue for the quarter, compared to 72% in the same period last year.

This shift reflects both the growth of our HPC business and the reality that without heavy reinvestment, mining market share naturally declines, and we are fine with that. While our hash rate stands to rebound in the second quarter, our primary focus remains on investing in our data center build-out and Cloud Services business. Turning to Cloud Services. Revenue for the segment increased 84% year-over-year and 14% sequentially to $14.8 million. Gross margins rebounded, expanding approximately 700 bps sequentially to 59%. We continue to expect segment margins to improve over time as revenue scales and as the impact from operating lease costs tied to our anchor customer contract is spread across a broader base. Based on our current contracted deployments, we expect stronger sequential revenue growth in the second quarter and continued growth in the third quarter of 2025.

Several deployments commenced midway through the first quarter, but we expect a recognition of full quarter of revenue contribution in the second quarter. Additionally, our initial deployment for DNA fund, a 576 H200 cluster began generating revenue in April and represents approximately $10 million in annualized revenue. In May, we expanded our relationship with DNA Fund through 2 new contracts totaling 616 H200 GPUs under two-year terms, representing approximately $10.8 million of additional annualized revenue. The expansion with this customer reflects our strategy of building trust through execution and using that as the foundation for expanding relationships over time. Our procurement strategy remains focused on aligning GPU purchases with contracted demand rather than taking speculative inventory risk.

We are effectively sold out of H200 capacity and have prioritized deployments backed by secured contracts. While overall demand for B200 GPUs remains healthy, uptake through our on-demand distribution partnership with Shadeform has been impacted by hardware reliability issues. We believe the crux of that issue is tied to the early iterations of servers we received, and we’re working with the OEM to address that issue. We expect this dynamic to improve over time as early hardware issues are hopefully resolved. We are currently marketing this cluster to customers for a multiyear contract. Separately, our anchor customer exercised their right to adjust the start date on their 464 B200 deployment from June 30 to August 20, the latest allowable date under the agreement.

As a result, this contract represents approximately $50 million of annualized revenue for 18 months. If we don’t secure an acceptable customer contract for our existing B200 cluster, we will likely use those GPUs to fulfill our anchor customer contract. We continue to prioritize securing multiyear deployments with creditworthy counterparties as the foundation for our growth strategy. Looking ahead, we’re engaged in several large contract discussions with a focus on securing multiyear agreements that are financeable and aligned with our capital efficiency objectives. Currently, we are conducting diligence and negotiating on 4 separate deployments with creditworthy counterparties. Each opportunity carries an annualized revenue potential above $100 million and a three to five-year contract term.

These are the types of contracts that we believe would support attractive financing structures, and we’re working on those financing options in parallel with negotiations. It’s too early to say whether we’ll ultimately win any of those deals, but we’re encouraged by the progress and the fact that we’re increasingly included in these processes. We believe this reflects the strength of the platform that we’ve built and are continuing to build through disciplined investment and execution. Finally, we are investing in proprietary software development to enhance our platform capabilities. A key milestone was the launch of our API layer for external provisioning of bare metal GPU servers with Shadeform as our first integration partner. This development not only expands our ability to integrate with third-party platforms, but also streamlines operations for our direct customers.

Over time, we expect that this will position WhiteFiber as a premium cloud infrastructure offering focused on maximum performance and reliability. Turning to colocation services with our data centers. Development activity continues across our sites. While the segment represents a small portion of Q1 revenue, we are laying the foundation for this segment to be a major growth engine in the coming years. At Montreal II, development time lines have shifted modestly, and we now expect initial capacity to come online around early to mid third quarter. The delay is largely due to the timing of debt financing, which we had expected to secure quicker, but we’re now in the very, very final stages. We recently completed the physical installation of a pilot GB200 liquid cool system at Montreal II.

While not yet operational, the project supports collaboration between our cloud services and colocation teams, providing our cloud team a platform to test next-generation hardware and our data center team early exposure to advanced liquid cooled systems. We secured our third data center site, Montreal III in April under our lease-to-own structure. Development remains on track with the Cerebras deployment expected to commence about two months from now. As a result, this is a build-to-suit deployment for a customer with a very sophisticated technology requirements, and we’re really proud to partner with them and help accelerate their growth plans. We are making progress on both sites and remain confident in our ability to meet key customer commitments.

In April, we disclosed via an 8-K filing that we signed a purchase agreement to acquire roughly 95-acre property in North Carolina, intended for data center development. This transaction remains subject to customary closing conditions, and we’re actively working through those processes. While it is too early to provide additional details, we are very excited about the potential strategic significance of the site of this for our platform, and we look forward to providing further updates as appropriate. In addition to our active projects, our broader development pipeline remains robust. We continue to pursue additional data center opportunities across Canada and the US with over 500 megawatts of potential capacity under evaluation or negotiation.

Customer demand for high-performance AI-optimized colocation remains strong, and we’re engaged in multiple active discussions that could drive incremental leasing at our existing and planned sites. On the financing side, we are now nearing finalization of a mortgage financing package for our Montreal II facility with a leading global banking partner. We expect to be in a position to announce the terms shortly. We believe this financing will validate the scalability and capital efficiency of our data center development model and provide a very strong foundation for future growth. Our data center platform is a critical pillar of our strategy to build a durable, diversified and high-margin infrastructure platform. We’re very excited about the opportunities ahead as we continue to expand capacity and deepen relationships with high-quality customers.

I’ll now hand over the line to Erke, who will discuss our financial results.

Erke Huang: Thank you, Sam. I will now walk through our financial results for the first quarter of 2025. Total revenue for the quarter was $25.1 million, a 17% decrease compared to the same quarter last year and slightly below the $26.1 million reported in the fourth quarter of 2024. The decline was primarily due to the lower Bitcoin mining revenue, which was partially offset by growth in our cloud services business and a full quarter of colocation revenue. Bitcoin mining revenue was $7.8 million, down 64% year-over-year and 26% sequentially. This decline reflects the impact of April 2024 happening, increased network difficulty and a temporary decrease in operational hash rate as we exited Coinmint facilities. We also retired and replaced older units [Technical Difficulty] optimize fleet efficiency.

Cloud services revenue was $14.8 million, an increase of 84% compared to the first quarter of 2024 and 14% sequentially. Growth was supported by new cloud contracts signed in both late 2024 and during the first quarter of 2025. [Technical Difficulty] contributed $1.6 million in revenue, up from $1.4 million in the fourth quarter. This reflects a full quarter of operations following our acquisition of Enovum in late 2024. Ethereum staking revenue was $0.6 million, slightly lower than the prior quarter. Our cost of revenue, excluding depreciation, was $12.8 million, down from $16.2 million in the same period last year and flat compared to the first quarter. This included $6.1 million in cloud services costs and $6.1 million in mining costs. Gross profit was $12.3 million, representing a total gross margin of 49%, that compares to 47% in the same quarter last year and 40% in the fourth quarter of 2024.

Cloud services gross margin expanded to 59% from 52% last quarter, reflecting improved utilization and scale. Colocation services gross margin improved modestly to 67%. General and administrative expenses were $8.2 million, up from $6 million in Q1 2024, which reflects an increase in headcount as we invest in our HPC business lines. Depreciation and amortization was $7.2 million, up slightly from the prior quarter, mainly due to a larger GPU fleet. As a note, we adjusted our depreciation schedule for cloud services for three years — from three years to five years, which we believe better reflects the useful lives of these assets. This accounting change decreased D&A by approximately $2.5 million for the quarter. Adjusted EBITDA was negative $44 million — $44.5 million compared to a positive $58.5 million in the first quarter of 2024.

This decline was primarily due to a $49.2 million mark-to-market loss on digital asset holdings, reflecting lower BTC and ETH prices at quarter-end. These were noncash charges and do not reflect changes in operating performance. GAAP net loss per share was $0.32 on a fully diluted basis compared to earnings of $0.43 per share in the first quarter of 2024. Now, turning to the balance sheet. As of March 31, we held $57.6 million in cash and cash equivalents and $3.7 million in restricted cash. The fair market value of our digital assets was approximately $80 million as of that date. Total liquidity, including digital assets and USDC was approximately $141 million as of that date. Since then, the price of Bitcoin has increased by 25% and the price of Ethereum has increased by 40%.

The market value of digital asset position has appreciated on a mark-to-market basis. Total assets were $485 million, and the shareholders’ equity was $417 million. We remain debt-free. CapEx for the quarter was $65 million, approximately $36 million for the CapEx was spent on the GPUs, including our B200 deployment. with the remainder spent on data center infrastructure, networking and storage equipment and Bitcoin mining units. I will now hand it back to Sam for closing remarks.

Sam Tabar: Thank you, Erke. Before opening the line for questions, I want to address our financing strategy and recent capital activity. We remain firmly committed to pursuing non-dilutive financing structures to support the expansion of our HPC platform. We’re actively working toward finalizing our first commercial mortgage financing, and we’re confident that this will provide a strong foundation for scaling our data center business efficiently. We have also initiated the process for commercial mortgage financing in the US. Regarding the filing of a new ATM registration, I want to emphasize that this was a mechanical renewal of our shelf capacity. It doesn’t reflect any change in our philosophy. We continue to view equity issuance as a tool to be used selectively and strategically with a strong preference for non-dilutive financing wherever possible.

During the first quarter, we raised approximately $10 million through our ATM program as part of our normal course of operations. Subsequent to quarter end, we raised approximately $48 million through the ATM. These proceeds strengthen our liquidity position and support certain strategic growth initiatives that we believe will be transformative for our business. We look forward to providing additional updates on these certain initiatives at the appropriate time. In parallel, we also sold approximately $32 million worth of Bitcoin holdings during the quarter to help fund growth while managing our use of equity issuance. We did not sell any ETH during the quarter, but transferred 3,400 into an internally managed fund, which we do not count as treasury holdings.

Maintaining a strong liquidity position is critical, not just for executing on our expansion plans, but also for building trust with our customers. Prospective colocation tenants view financial strength as a key factor in selecting infrastructure partners, and our balance sheet reinforces our ability to deliver projects reliably. Our focus remains on executing our growth strategy, scaling our infrastructure platform and pursuing disciplined shareholder-friendly capital deployment. As our business continues to evolve, we are actively evaluating corporate structure and strategic initiatives to maximize long-term shareholder value. With that, I would like to turn the call over to the operator for Q&A. But as a note, we have Billy Krassakopoulos, who leads our data center business; and Ben Lamson, Head of Revenue for our Cloud business on the line for Q&A.

Operator: [Operator Instructions] we can take our first question from George Sutton with Craig-Hallum.

George Sutton: Thank you. Ben, I’d like to address my question to you, if I could. On the WhiteFiber rebranding, can you just give us an update of how that’s been received in the market? And then I know you’re working on a lot of platform initiatives that are somewhat new to the market. Can you just give us any update there?

Ben Lamson: Yeah, happy to. So the rebrand has been really well received. We actually have gone through a few iterations on the website as we’ve gotten feedback from customers and partners. You may have seen we’ve recently launched a new version of the website about two weeks ago, 1.5 weeks ago, and the reviews have been fantastic. So really, really happy with how that’s been panning out for us. In terms of on the platform layer, I want to be careful not release anything a little too soon. We will have some news coming out here in the next couple of weeks that will be really exciting, some first-to-market technology that we’re going to be releasing. We’re just waiting on some independent third parties to publish some benchmarks around that.

So look out for that. That’s going to be coming in the coming weeks. And then we’ve got some other developments coming later this year on the platform side around cross data center workloads, which we believe is going to be really a revolutionary technology in terms of us being first to market to productize that. So I don’t want to say too much just because I want to allow some of these independent third parties to release some of this data, and we’ll be able to leverage that media flash there appropriately. But just stay tuned in the next few weeks for some announcements.

George Sutton: Understand. Thank you.

Operator: Thank you. Our next question comes from Brian Dobson with Clear Street.

Brian Dobson: Thanks very much for taking the question this morning. Do you think that we could take maybe a step back and you could provide a 30,000-foot view on how you see demand from hyperscalers and enterprise users evolving over the next six months or so?

Sam Tabar: Would that be for the data center side? If so, Billy, please feel free to weigh in. Just to clarify, that’s with respect to data center colocation, correct?

Brian Dobson: Yeah, that’s right.

Sam Tabar: Billy, would you like to take that?

Billy Krassakopoulos: We’re seeing very strong and positive demand from not only hyperscalers, but medium-sized neo clouds as well for capacity that we’re evaluating and looking to come online later on this year. We should have some news in the next couple of months about that.

Brian Dobson: That’s very exciting. Thank you very much.

Operator: Thank you. Our next question comes from Mike Grondahl with Northland Securities.

Mike Grondahl: Hey, guys. Good morning. I think what you were describing is a delay for [customer one] (ph) from June 30 to August 20. One, can you talk a little bit about why — and then I think you were saying how you were going to use maybe those GPUs for your own book. Just kind of walk us through the options you have there.

Sam Tabar: Yeah. I think the maximum allowable date is the date they’ve exercised. And as you mentioned, Mike, we have the cluster to honor what they’re looking for. So that’s a good thing since we already have that cluster, and we could just use that if we don’t score multiyear contracts for that current inventory. Is your question about why they shifted?

Mike Grondahl: And then — so what you’re saying is, hey, the GPUs might not sit idle until August if you can put them in a new customer win?

Sam Tabar: Correct.

Mike Grondahl: Got it.

Sam Tabar: But as for the reasons for shifting, I’m not — Erke, do you have any color on the reasons they’re shifting the start date for two months?

Erke Huang: I think we have some internal product development schedule change. So I’d like to — they still want those compute, but they’d like to exercise the option to extend at a little later date. And we have installed and deployed those equipment in our Atlantic data center already. So right now, we can sell those computes, and I know Ben is working on that. There’s a few counterparties we’re negotiating with for multiyear contracts. And on top of that, we are putting this B200 compute on-demand — Shadeform as well. So it’s already generating some revenue.

Mike Grondahl: Okay. Great. Thank you.

Operator: Thank you. Our next question comes from Nick Giles with B. Riley Securities.

Sam Tabar: Hi, Nick.

Nick Giles: Thank you, operator. Good morning, everyone. Guys, it’s good to see you further expand into the US with the North Carolina agreement. So my first question is, how should we think about your desire to continue that expansion in the US versus Canada? And secondly, apologies if I missed this, but can you just outline when those megawatts could be available, what the ramp will ultimately look like and CapEx expectations?

Sam Tabar: Yeah. Well, look, as you know, we disclosed in an 8-K that we signed a purchase agreement to acquire a roughly 95-acre property in North Carolina. The transaction remains subject to some closing conditions. So it’s too early for us to provide additional detail until it closes. But I can tell you, I’m deeply excited about its potential strategic significance. And we will definitely update you and the markets as soon as if this closes. The deal isn’t closed. So there isn’t much we can say or frankly, or what we want to say until something is finalized in either direction. But to your question about the broader colocation development pipeline, we’re evaluating and negotiating over 500 megawatts of additional potential capacity across Canada and the U.S. These sites range from small sites in the 5 to 20-megawatt range to sites with over 100 megawatts.

And we continue to target those sites in adjacent to market cities or in proven corridors of customer demand. That’s really important for us geographically. We focus on retrofits. So, we look at sites that aren’t currently configured as data centers, but would lend themselves well to redevelopment and reduce costs and time lines relative to a greenfield site. And the team that we have, when we acquired Enovum last year, it wasn’t just about an acquisition of a Tier-3 data center. But more importantly, it was about a team that’s been doing this their entire careers and also a huge pipeline that they’ve identified through LOIs and through a lot of work, about 500 megawatts worth. So they — the secret sauce with respect to that particular acquisition was their experience in retrofit and how they could do it faster and cheaper, and they have a track record of doing it faster and cheaper.

And this is really the secret sauce of we’re not trying to sort of figure it out as we go. We have acquired a team that’s been doing this for a long time before this sector got hot. So we’re really looking forward to developing that colocation pipeline and there will be some exciting — definitely some exciting announcements in the medium-term future, if not the near-term future.

Nick Giles: Sam, I appreciate all the color and look forward to the updates.

Sam Tabar: Likewise. Thank you for asking.

Operator: Thank you. Our next question comes from Joe Gomes with Noble Capital.

Joe Gomes: Good morning. Just wanted to kind of get your thoughts, if I look through the 10-Q this morning and you’ve got $33 million on the balance sheet of investments and at the same time you’re raising equity. And just trying to get your thought on how do you weigh that? Is it more — is it a better opportunity to liquidate those investments and not have to sell so many shares in the ATMs? Just trying get your thought process on that.

Sam Tabar: Yeah, absolutely. So the recent filing was a mechanical renewal just to maintain flexibility and optionality, it doesn’t reflect the change in our posture towards selective and strategic use of equity issuance. We understand — I completely understand the optics are not great, given the size of the ATM relative to our market cap. The market cap is transitory having the ATM on file is really just a tool we have access to. It’s worth more to have that optionality in future years than to forego in favor of optics. But look, liquidity is critical for executing strategic initiatives and building customer confidence as mentioned on my — on the call earlier. We do balance raising equity with selling digital assets to fund growth responsibly.

We did sell some Bitcoin. Fortunately, we did not sell any ETH, as you know that ETH has had a really great run. And we’re still pretty strong believers. There’s a lot of juice in there. But we also are very excited to announce mortgage financing which is really inexpensive capital in order to fund the growth of our data center business, and we look forward to announcing the terms of that mortgage financing very soon. Then that will really help with respect to how we fund our growth. We want to use cheap sources of financing. We do not want to use the ATM. But it has to be an option.

Joe Gomes: Thanks.

Operator: [Operator Instructions] Our next question comes from Kevin Dede with H.C. Wainwright.

Sam Tabar: Hi, Kevin.

Kevin Dede: Good morning, Sam, Erke. Thanks for having me on the call. Apart — Erke mentioned changing the depreciation schedule. But apart from that, Sam, can you walk through the levers that you can pull or the variables that you see changing the gross margin profile of your cloud and colo business and maybe layer in a little discussion on the GPU procurement strategy and filling the Shadeform and DNA fund business?

Sam Tabar: There’s a few questions there. Let me rephrase a couple of those questions. With respect to — I’m not sure if it was a question or a comment about changing the depreciation schedule. But three years was just overly conservative and way too aggressive. If you look at others in the industry, they do five, some even do more than five. So even being at five at this point is still, we believe, responsible and conservative. I think the next question you had in there was what are the different levers to increase the margins of our cloud business and our data center business. Is that what I — is that correct?

Kevin Dede: Yes, absolutely, Sam. Thank you very much. I apologize for the connection that I’m working with here.

Sam Tabar: Yeah. No, all good. I mean, besides charging more, of course, to the customers, I’d love for Billy to weigh in on the different levers to increase margins. Go ahead, Cameron.

Cameron Schnier: I can just jump in, Sam. I mean, Kevin, the biggest thing on the cloud side is really just spreading the operating lease for our anchor customer over a larger revenue base because that’s one of the major COGS items right now. and it’s effectively a financing structure, but it’s above the line. So just getting more revenue, the absorption across that will naturally drive gross margin up. So any increase in revenue is broadly gross margin accretive. Billy can add in on the data center side, but I think the way we look at it is pretty much every leading-edge contract that we sign and revenue that goes on will lift gross margins as well.

Billy Krassakopoulos: On the data center side, margins are very predictable and can be relied on long-term contracts with clients who are very sticky, give us great insight on what the numbers will be over the next couple of years. We’re aiming for minimum five-year contracts on that segment. So the predictability and the reliability of those are very certain.

Kevin Dede: Yeah, Sam, could you just touch on your procurement thinking, your GPU procurement thinking? I mean I know in the past, you both bought and leased. I’m just wondering if that thinking and philosophy is altered in any way.

Sam Tabar: I think Erke is closer to that work stream than I am. So I’d love to — the procurement for — on the cloud side, just to clarify, which side of the business are you referring to? Procurement for equipment for data centers or procurement for the GPU.

Kevin Dede: Yeah, GPU for cloud, right? Because you’re — digital is responsible for providing the compute for both Shadeform and DNA, at least as I understand.

Sam Tabar: Yeah, I’ll pass that question to Ben or Erke.

Erke Huang: We still remain the same strategy, try to sign contracts and tied to a procurement so that we minimize our exposure to speculative procurement. But at the same time, we did procure some GPUs prior of signing a definitive agreement that would allow us to do some R&D and do some benchmarking and testing as well. So we had completed our deployments in Iceland. Right now, there’s a couple of deployments we’re working on in Canada at different sites as well. And at a higher level, if I may add, we tend to procure those most advanced technologies or GPUs, which are more attractive to end users at early stage as well. And we have those relationships with OEMs and NVIDIA to get those chips at a faster or more accelerated time line.

Kevin Dede: Great. Thank you. Appreciate it. Thanks, Sam.

Sam Tabar: Yeah. Thank you.

Operator: Thank you. It appears that we have no further questions at this time. Mr. Tabar, I’ll hand the call back to you for closing remarks.

Sam Tabar: Okay. Well, then that’s it. And thank you for joining us on the call today, everybody. We appreciate your continued interest and support. We look forward to speaking with you again in the next quarter. This officially concludes our call, and have a great day, everybody.

Operator: This concludes today’s call. Thank you for your participation. You may now disconnect.