BitFuFu Inc.

I would now like to turn the call over to your first speaker today, Mr. Charlie Brady. Please go ahead.

Thank you, operator. Ladies and gentlemen, good day and welcome to BitFuFu's full year 2025 earnings conference call. The company's financial results were released earlier today and are available on BitFuFu's investor relations website at irbitfufu.com and globe-newswire.com. Joining me today on the call are Chairman and CEO Leo Liu and CFO Kala Jow. Before we begin, please note that today's discussion will contain forward-looking statements made pursuant to the safe harbor provisions of the US Private Securities Litigation Reform Act of 1995. Statements that are not historical facts, including statements about the company's beliefs and expectations, are forward-looking statements. Forward-looking statements involve risks and uncertainties that could cause actual results to differ materially from management's current expectations. Potential risks and uncertainties include, but are not limited to, those outlined in the company's public filings with the U.S. Securities and Exchange Commission. The company assumes no obligation to update any forward-looking statements, except as required by applicable law. We will discuss non-GAAP financial information on this call. The company provides this information to supplement information prepared in accordance with U.S. generally accepted accounting principles. A reconciliation of these measures to the company's reported GAAP results can be found in the reconciliation tables provided in today's earnings release. Finally, it's important to note that while we won't be conducting Q&A on this call, you can email your questions to ir at bitfufu.com and we'll respond as quickly as possible, typically within 24 hours. I will now turn the call over to Leo Liu, Chairman and CEO of the company.

Thanks, Charlie, and thank you all for joining us today. 2025 was a challenging year for the cryptocurrency industry, particularly for mining enterprises. When the price of Bitcoin surged rapidly during the second and third quarters, hitting an all-time high of $126,000, market sentiment leaned heavily toward targets of $150,000 or even $200,000 by year end. Consequently, many market participants increased capital expenditures and leverage, while others continued to accumulate more Bitcoins. Bitcoin never follows a script. After the run-up, it dropped below $100,000 in November, below $90,000 in December, and entered a weaker market in 2026. That volatility created severe cash flow pressure for some market participants and forced many liquidations. Looking back, Bitfufu executed its strategic plan and maintained rigorous operational discipline throughout 2025. We did not lever up or chase unprofitable growth at the top of the cycle, and we preserved liquidity so we would not need to sell Bitcoin in weak markets to support operations or reduce leverage. We achieved the majority of the goals set at the beginning of 2025. and built a solid foundation to navigate the current weaker market conditions. I will touch on a few of these points in a moment. As reported in our earnings release earlier today, for the full year 2025, we generated total revenue of $475.6 million. Our cloud mining business continued to be our largest revenue contributor, generating a record $350.6 million, up over 29% versus 2024, and accounting for nearly 74% of our total revenue for the year. Kala will discuss this and our other revenue lines in more detail later. Operationally, we continued executing our hybrid model of cloud mining and self-mining, along with complementary mining services such as hosting and miner sales, mining pool services and our proprietary mining rig operating system. We are pleased to see the powerful synergies between these business lines. For instance, our US deployed mining facilities allow us to offer highly competitive hosting rates and power solutions. This, in turn, attracts customers to purchase hardware directly from us, effectively driving our minor sales activity. Furthermore, the robust Cash flow generated from our miner sales and cloud mining solutions provides the necessary liquidity to support our self-mining operations and Bitcoin accumulation treasury strategy. In 2025, we also successfully onboarded additional suppliers and diversified our hash rate supply, effectively mitigating potential risks associated with supplier concentration. We continuously optimize our fleet and site mix, and we ended December 2025 with 26.1 EXA hash of managed hash rate, including 3.7 EXA hash of self-owned hash rate, and 22.4 EXA hash from third-party suppliers and hosting customers. Total available power capacity was 478 megawatts, including 164 megawatts of control capacity in Ethiopia and the US. We chose to stay disciplined regarding capital expenditure on hardware. Instead of expansion for expansion's sake, we align our hardware procurement with the addition of new self-controlled power capacity, keeping our production costs highly competitive. By balancing our usage of power capacity between self-mining and third-party hosting, We capture both the upside in Bitcoin and the consistent profitability of hosting service fees. This philosophy is similar as our broader cloud mining and self-mining hybrid model. In this business, success comes down to precise ROI decisions. In 2025, we focused on execution at the site and fleet level, improving efficiency and reliability, that operating focus gives us a stronger foundation to manage through the current weaker market conditions. Now, let me shift gears for a moment and discuss how Bitfufu differentiates itself compared to other miners. This is also a question that we are frequently asked by investors. We believe Bitfufu stands out because we operate like a services and platform business, not a single product miner. Our platform combines a hybrid operating model of cloud mining and self mining, a purpose-built hashrate management system we call Aladdin, and an integrated suite of services that serves both individual users and large-scale mining operators. First, our hybrid model balances upside and resilience. The mix of self mining and cloud mining allows us to participate in Bitcoin's upside by increasing our treasury holdings while also building recurring service-based revenue with better cash flow visibility across market cycles. We believe the top line growth of our cloud mining business and its greater contribution to total revenue reflects sustained customer demand and durability, not a one-off trend. Second, our proprietary Aladdin hash rate management system is built to optimize our fleet, and deliver hash rate reliably. Aladdin dispatches and monitors hash rate across dozens of sites and hundreds of thousands of machines with real-time data, predictive maintenance, and flexible allocation. The practical advantage is reliability. We can reroute capacity and keep performance stable when conditions change. Aladdin is designed to scale to manage millions of machines. We have an integrated platform with real adoption. Beyond mining, we provide hosting services, miner sales, mining pool services through Bitfufu Pool, and our own mining software, Bitfufu OS. By late 2025, our registered cloud mining users surpassed 675,000, while the majority of our cloud mining revenue continues to be generated from institutional customers. We believe there is enormous potential for retail customer participation. Finally, we believe governance and supply chain credibility are competitive advantages. We are a NASDAQ listed company subject to stringent compliance requirements. We also operate with KYC IML controls and user level transparency that institutional customers expect. This transparency helps build trust and is one reason we have built a leading position in cloud mining and generated a net dollar retention rate of 100% in 2025. In the current market environment, we have witnessed many miners prioritizing near-term liquidity by selling Bitcoin holdings to support operations or reduce significant leverage positions. Our approach is different. We manage liquidity conservatively and seek to build our Bitcoin treasury through the cycle while maintaining flexibility. We also maintain a conservative debt profile to provide security during low points in the market cycle. As of December 31st, 2025, we held 1,778 Bitcoins compared to 1,720 Bitcoins at the end of 2024. As we entered 2026, we continued stacking. reaching 1,830 Bitcoins as of February 28th, 2026. Importantly, at year-end, we reduced pledged Bitcoins to 274 Bitcoins from 633 Bitcoins as of year-end 2024, increasing the Bitcoins available on our balance sheet and improving financial flexibility. Looking ahead to 2026, I believe our growth opportunities will stem from several key areas. First, regarding the sources of hash rate expansion, in a bear market, many miners face cash flow pressures and may seek to liquidate hash rate to recover capital for debt repayment or fleet upgrades. Similarly, hardware manufacturers may prefer to sell short-term hash rate to alleviate inventory pressure. The prevailing hash rate supply logic shifts from pursuing high alpha returns to asset preservation through extreme operational efficiency. Consequently, the hash rate supply can actually become more active in a bear market than in a bull market. In terms of market demand, institutional clients who purchased cloud hash rate at higher price levels often increase their positions during market dips to lower their average cost per coin. We also anticipate the entry of new customers with higher risk appetites. who recognize the value of entering the market during periods of low activity to capture greater future returns. Due to the inherent leverage effect of cloud mining solutions, customers can typically accumulate more Bitcoins compared to direct exchange purchases. In 2026, Bitfufu remains strategically focused on acquiring infrastructure, and we are continuously evaluating potential partnership opportunities. This is the core of our vertical integration strategy. We believe that power capacity is an appreciating asset that not only lowers our self-mining costs and enhances competitiveness, but also provides a robust foundation for our minor sales and hosting businesses. Our priorities in 2026 are to, firstly, scale cloud mining and maintain strong customer retention. Expand managed capacity across hash rate and power while maintaining disciplined returns. Thirdly, continue improving reliability and uptime through Aladdin. Fourthly, optimize capital allocation and liquidity. And fifthly, continue building our Bitcoin treasury opportunistically. I will now turn the call over to Kala to provide more details on our financial results.

Thank you, Leo. Good morning, everyone. Over the past year, our focus on operational execution at the site and fleet level enhanced our efficiency and reliability. While we recorded the net loss, we delivered positive adjusted EBITDA. Additionally, we achieved revenue growth and preserved a healthy balance sheet. Now, I'll walk through our full year 2025 financial and operating results. The total revenue for the full year 2025 was $475.8 million, compared with $463.3 million in 2024. The year-over-year increase was primarily driven by growth in cloud mining solutions, mining equipment sales, and hosting services, partially offset by a decline in self-mining revenue due to the lower hash price and increased network difficulty. Cloud mining solutions revenue in 2025 grew 29.3% year over year to $350.6 million compared with $271 million in 2024. Cloud mining solutions represented approximately 74% of revenue in 2025 compared to approximately 59% in 2024. demand for cloud mining solutions remained strong and continued to exceed available supply. As a result, we were able to improve pricing and redirected the majority of our available hash rate to institutional customers at higher price points. In 2025, existing customers represented 79% of cloud mining solution sales and new customers represented 21%, compared with 77% and 23%. respectively in 2024. Net dollar retention was 100% in 2025 versus 117% in 2024. Self-mining revenue was $63.1 million versus $157.5 million in 2024 and accounted for about 13% of total revenue compared to 34% in 2024. In 2025, we allocated 43% of our owned hash rate to cloud mining and 57% to self-mining. For least hash rate, 91% supported cloud mining and 9% supported self-mining. This reflects a meaningful shift from 2024 when only 3% of owned hash rate supported cloud mining and 73% of least capacity supported cloud mining Year over year, the allocation of owned hashrate to cloud mining increased by 40 percentage points, driven by the strength in demand for cloud mining solutions and our ability to flex capacity toward customer demand. We also increased the share of leased capacity directed to cloud mining by 18 percentage points, supporting that same demand-driven shift. Turning to hashrate sourcing, in 2025, 8% of cloud mining hashrate was supplied by our owned miners and 92% from purchase or lease hash rate from third parties. This compares to 2024 when cloud mining was supported entirely by third party capacity. For self mining, hash rate from owned miners increased to in 2025 from 24% in 2024. For the full year 2025, the average cash cost to mine one Bitcoin was $77,573. This reflects a blended cost structure across our self-mining operations, with approximately $61,000 per Bitcoin from owned machines and approximately $98,000 per Bitcoin from leased hash rate. The use of leased hash rate, which typically carries contract durations of 3 to 12 months, provides operational flexibility. This allows capacity to be periodically rebalanced and repriced in line with evolving market conditions. Given that least pricing is generally linked to prevailing hash price, this approach enables a more dynamic cost management framework, particularly during periods of market volatility. While least capacity carries a higher nominal cost, its shorter duration reduces long-term capital commitment and mitigates risks associated with hardware obsolescence and depreciation. As a result, A hybrid model combining owned infrastructure and lease capacity is maintained, supporting both cost efficiency over the long term and flexibility to adapt to changing market conditions. The key takeaway is that we still have meaningly opportunity to lower our cost to mine Bitcoin as we still lease a majority of mining capacity. Mining equipment sales increased 76.4% year over year to $53.7 million. versus $30.5 million in the prior year. Around 60% of the equipment that we sold in 2025 was S21 series. Full year revenue from hosting services and other was $8.4 million versus $4.3 million in 2024. Growth in this area was driven by our minor with hosting program that was launched in the second half of 2025. Full year 2025 growth profit after deducting depreciation and amortization was $26.7 million, resulting in a gross margin of 5.7% compared to 6.4% for full year 2024. Higher gross margins in cloud mining and hosting services were offset by lower margins in self-mining and equipment sales. Before deducting depreciation and amortization, 2025 gross profit margin was 11.6% versus 11.7% in 2024. For 2025, sales and marketing expense was $3.7 million. General and administrative expense was $9.1 million. And research and development expense was $2.5 million. For 2024, those line items were $7.5 million, $25.3 million, and $5.6 million respectively. The primary driver of the year-over-year decline was significantly lower 2025 stock compensation expense. Full-year 2025 net income was a loss of $57.4 million compared with net income of $54 million in 2024. The year-over-year decline was due to a larger non-cash impairment loss of mining equipment increased unrealized fair value losses on digital assets, receivables, and payables in Bitcoin, and a lower realized gain on the sale of digital assets. Adjusted EBITDA for the full year was $8.3 million. Turning to our balance sheet, we ended December 31st, 2025, with cash and cash equivalents of $27.8 million and digital assets of $149.3 million compared with $45.1 million and $129.9 million, respectively, at year-end 2024. In addition, as of December 31st, 2025, the company held $24.1 million in digital asset collateral receivables, which represents Bitcoin pledged to lenders in exchange for borrowings. As Leo noted, we ended the year with 1,778 Bitcoins and continued building into 2026. As of February 28, 2026, we held 1,830 Bitcoins. Looking ahead to 2026, we are evaluating opportunities to expand our owned capacity footprint, particularly in North America and the Middle East. We will update the market if and when we sign definitive agreements. In terms of our mining fleet, our objective for 2026 is to gradually upgrade our self-owned S19 series miners to newer generation machines with greater efficiency. In a bear market, asset valuations often reset to more attractive levels, which can create compelling opportunities. At the same time, as next generation miners enter the market, manufacturers may discount legacy models. we will strategically evaluate the economic benefits of both options and prioritize the one with the best returns and payback, rather than simply pursuing the newest technology. Separately, expanding our own power and infrastructure remains a strategic focus for 2026. The company continues to have sufficient liquidity to meet working capital needs. As of the end of February 2026, the company had $85 million available under our 100 million revolving credit facility. We remain focused on disciplined capital allocation, maintaining liquidity, and optimizing our funding structure to preserve flexibility through the cycle. Before I turn the call back to Leo for his closing remarks, I want to address a question we have received from analysts and investors. why some customers are willing to pay a premium for cloud mining services instead of simply buying Bitcoin directly in the spot market. There are a few clear drivers behind this. First, dollar cost averaging and cost visibility. Cloud mining allows customers to build Bitcoin exposure over time through a more systematic approach. By placing cloud mining orders consistently across different market conditions, customers can effectively implement a managed dollar cost averaging strategy, reducing the risk of trying to time the spot market. Second, financial leverage through installment payments. The installment structures in cloud mining solutions create built-in leverage by deferring a significant portion of the service fee component. Because the service fee typically represents about 70% to 75% of the total contract cost, the deferral allows more upfront investment to go toward purchasing hash rate that mines Bitcoin. This increases hash rate exposure with lower initial liquidity. Third, upside optionality on network dynamics. Unlike the fixed outcome of buying Bitcoins on an exchange, cloud mining output is dynamically adjusted based on network difficulty and transaction fees. This introduces strategic uncertainty with significant upside. When network difficulty drops or on-chain transactions up to the surge, clients can potentially earn substantially more Bitcoins than they would have through a one-time market purchase. Fourth, financial reporting and compliance. For some institutional customers, the accounting treatment is a key driver. Cloud mining allows them to record operational revenue from mining activities rather than reporting investment from direct Bitcoin purchasing and holdings. Furthermore, cloud mining mitigates counterparty and trading risks. The Bitcoin produced is categorized as newly minted coins, which are untainted by prior transaction history, ensuring full compliance with the stringent AML and audit requirements of institutional investors. And fifth, supply-demand dynamics and the value of reliability at scale. To our knowledge, there are limited sources for transparent, verifiable, large-scale cloud mining at public company standards of governance and disclosure. When demand for reliable mining exposure exceeds available hash rate supply, large-scale customers are often willing to pay a premium to secure consistent delivery and high-volume production that can be difficult to source in the open market. With that, I will now turn the call back over to Leo.

Thanks, Gala. In summary, we're building Bitfufu to perform across market cycles. Cloud mining remains a key differentiator, enabling recurring service-based revenue alongside our self-mining exposure, and our integrated platform broadens how customers participate in mining. We are also continuing to build our Bitcoin treasury at a time when many miners are selling, while staying focused on disciplined capital allocation and liquidity.

We appreciate your time today and your continued interest in Bichu Fufu.