TE - NYSE

Good day, and thank you for standing by. Welcome to the T1 Energy Third Quarter 2025 Earnings Conference Call. [Operator Instructions] Please be advised that today's conference is being recorded. I would now like to hand the conference over to your speaker today, Jeffrey Spittel, Executive Vice President, Investor Relations and Corporate Development. Please go ahead.

Good morning, and welcome to T1 Energy's Third Quarter 2025 Earnings Conference Call. With me today on the call are Dan Barcelo, our Chief Executive Officer and Chairman of the Board; Evan Calio, our Chief Financial Officer; Jaime Gualy, our Chief Operating Officer; and Otto Erster Bergesen, our SVP of Project Development. During today's call, management may make forward-looking statements about our business. These forward-looking statements involve significant risks and uncertainties that could cause actual results to differ materially from expectations. Most of these factors are outside T1's control and are difficult to predict. Additional information about risk factors that could materially affect our business is available in our annual report on Form 10-K filed with the Securities and Exchange Commission and our other filings made with the SEC, all of which are available on the Investor Relations section of our website. With that, I'll turn the call over to Dan.

Thanks, Jeff, and welcome, everyone, to our third quarter earnings call. Let's turn to Slide 4, please. Many of you may be new to the T1 story this quarter, so we'll begin today with a brief look at our current position in the U.S. solar market. With 5 gigawatts of annual capacity at G1_Dallas, T1 is the largest American manufacturer of silicon-based solar modules, and we are the second largest American-owned solar module producer in the U.S., but we're just getting started. As we'll discuss on today's call, we are advancing our plan to start construction of the first 2.1 gigawatt phase of our U.S. solar cell fab, G2_Austin, before year-end. G2 is the centerpiece of our strategy to build the first end-to-end domestic polysilicon solar supply chain in the U.S. This strategy is intended to competitively differentiate T1 and to align the company with the growth dynamics in U.S. power markets. Now let's move to Slide 5 for a closer look at the big picture developments, which underpin our strategy. Today's theme is powering America. With U.S. electricity demand growing faster than it has in decades, we are positioning T1 as a homegrown enabler of 3 increasingly evident macro trends: accelerating U.S. AI development, onshoring of advanced American manufacturing and strengthening American energy security. These 3 trends are the thematic pillars of T1's investors' case. Energy is key to unlocking the future of AI. New data centers now routinely require gigawatts of electricity, and they are growing exponentially more compute and energy intensive. Energy has emerged as the leading checkpoint for AI growth. The U.S. has the natural resources and talent to debottleneck the AI equation, and T1 plans to contribute by bringing the capability to produce leading-edge solar technology at scale domestically. T1 intends to power American AI by investing in American advanced manufacturing. The reshoring of manufacturing is another trend that is driving electricity demand growth and presenting T1 with the opportunity to strengthen critical U.S. energy supply chains. We have ramped up domestic PV module production in G1_Dallas. We are advancing towards the expected start of construction at G2_Austin, our U.S. solar cell fab, and we are expanding our U.S. supply chain through our recently announced partnerships with Hemlock/Corning, Nextpower and Talon PV. We have entered an era when control of digital intelligence and AI infrastructure will determine the fate of nations. This underscores the strategic value of domestic energy capacity, and we believe T1's plan to build a domestic PV solar supply chain will contribute to U.S. energy security. In addition, standing up a domestic end-to-end polysilicon supply chain should strengthen our national ability to produce semiconductors, advanced materials and grid and space technologies, all of which involve common inputs and production processes. Turning to Slide 6. Let's drill down into the AI power theme. If the U.S. is to maintain its lead in AI, we need more electrons and we need them now. Leaders from the technology industry have suggested the U.S. must double the 2024 pace of electricity additions to 100 gigawatts per year to close the widening electron chasm between AI-driven demand and power availability. At T1, we are proponents of U.S. energy abundance, and we endorse the strategic merits of adding new natural gas and nuclear power capacity to our grid, but those technologies can only play a limited role in the near term due to swollen order backlogs, permitting red tape and construction cycle times for new generation facilities. Solar, coupled with battery storage, is the obvious choice to bridge this gap as a rapidly deployable resource at scale. The dawn of the AI age is a company-making opportunity for T1. We have available capacity at G1_Dallas, where we recently eclipsed the daily production record equating to an annualized rate of 5.2 gigawatts. As we look to 2026 and beyond, our plans to integrate upstream of G1 will position T1 as the first company that can offer hyperscalers and their partners a high domestic content, polysilicon-based TOPCon solar module. Now let's move to Slide 7 for an update on T1's business. Shortly after we announced our preliminary third quarter results in October, we closed 2 successful equity capital markets transactions. T1 raised $72 million in gross proceeds from a registered direct common equity offering with high-quality new and existing institutional equity investors. And as previously disclosed, T1 entered a $100 million commitment for the issuance of preferred and common stock to certain funds and accounts managed by Encompass Capital Advisors, LLC in connection with T1's acquisition of Trina Solar's U.S. manufacturing assets. Last month, T1 elected to make the second and final draw of $50 million pursuant to this $100 million commitment. This infusion of equity capital positions T1 to begin the first phase of construction at G2_Austin during the fourth quarter of 2025. Although we initially intended to focus on raising debt prior to an equity tranche to partially fund the first phase of construction at G2_Austin, these 2 transactions enable us to raise capital at attractive terms while we engage with prospective debt investors and advance the traditional project financing. The additional trading liquidity from a higher share count and market capitalization also provides opportunities for us to add new shareholders who were previously unable to trade in our stock. At T1, we are focused on shareholder value and as equity owners ourselves, we are highly sensitive to dilution. So we'll continue to use equity judiciously to fund growth CapEx while we optimize our capital stack. Our capital formation progress positions us to add G2 to our expanding domestic polysilicon solar supply chain, which now encompasses a growing network of American partners. In August, we announced an expanded polysilicon supply agreement to include production of American-made solar wafers with Hemlock/Corning. And in October, we signed a framework agreement with Nextpower for the provision of domestic steel frames, and we made a strategic minority equity investment in Talon PV LLC, which is building a U.S. solar cell fab in Texas. These partnerships are foundational to T1's mission to build the first integrated American polysilicon solar supply chain. Our expanding partnership network and the domestication of our supply chain are also key elements of T1's policy playbook. As we highlighted on the second quarter call, our team continues to advance the de-FEOCing process to maintain T1's eligibility for Section 45X tax credits in 2026 and beyond due to requirements in the OBBB. Moreover, our commitment to invest in advanced American manufacturing and critical domestic energy supply chains are consistent with some of the administration's top priorities. Turning to operations. We continue to ramp production and sales during the third quarter at G1_Dallas, our state-of-the-art solar module facility. During the fourth quarter, we expect to generate significantly higher sales and EBITDA as we ship modules under previously booked merchant sales agreements and as we sell down inventory to customers who are clearing out 45X eligible modules before year-end. As a result, our 2025 EBITDA guidance of $25 million to $50 million is unchanged. While we build our business in the U.S., we continue to advance our goal to generate value from our legacy European assets, which are attracting interest for repurposed data center applications. We look forward to providing updates on this initiative as warranted by our progress. As we do on each quarterly earnings call, we have a rotating guest speaker from T1's management team to expand on an important topic. Since this quarter's theme is Powering America, I'd like to introduce our SVP of Project Development, Otto Erster Bergesen, to provide an update on G2_Austin, which will be the centerpiece of T1's domestic supply chain and where we are approaching the start of construction. Otto?

Thank you, Dan. Let's turn to Slide 8. After months of work, we have a great design developed and Tier 1 partners contracted to help us move ahead with G2_Austin. We are ready to enter full execution shortly. We're pursuing a 2-phased approach to reach more than 5 gigawatts of capacity of solar cell manufacturing. Phase 1 will be a 2.1 gigawatt fab, which we plan to follow with a 3.2 gigawatt Phase 2. If offtake level permits, we can expand the second phase. The basis of design is Trina Solar's more than 100 gigawatts of solar cell fabs in general and their 5-gigawatt state-of-the-art Huai’'an fab in particular. We have customized this design together with JFE Engineering in China and later with SSOE as our U.S. engineering firm. We have been working very closely with Trina, JFE, SSOE and other companies over the past 10 months to leverage their project and operational experience while securing U.S. compliance and tailoring to U.S. conditions. Yates Construction has been selected as our general contractor. We have worked with Yates since May to provide preconstruction services, focusing on constructability and engagement of global and local subcontractors. Laplace has been selected as our EPC turnkey partner for the production line equipment. In August, we began working with Laplace on detailed design and preparations for equipment manufacturing. Laplace was a first mover on TOPCon and has extensive experience in the TOPCon space. They have been part of solar cell fabs for more than 400 gigawatts of capacity. T1 has great confidence in their ability to deliver top quality and to achieve according to their performance guarantee under the contract. The past few months, we've been working closely with Laplace and Yates to engage critical subcontractors to identify and address long lead items. We are pleased to report that the project has been very well received in the market and that we are currently contracting with subs to support the project schedule. For example, we have secured a very beneficial mill roll contract that enabled us to start erecting steel in March 2026. We have also secured favorable terms on long-lead electrical equipment like switchgears, generators and transformers. Finally, we have built a strong team, combining Tier 1 partners with a solid in-house project management and engineering team. If you take one thing from our portion of today's presentation, I want it to be that we have a world-class team with the experience and technical expertise to execute the G2_Austin project successfully, and we look forward to breaking ground before year-end. With that, I'll turn it back over to Dan.

Thanks, Otto. Let's turn to Slide 9. While we move towards the expected start of construction at G2, production and sales continue to ramp at G1, our state-of-the-art U.S. module facility. We have produced more than 2.2 gigawatts of modules year-to-date, and we are on track to meet our unchanged 2025 production plan of 2.6 to 3 gigawatts. And in October, we achieved a daily production record of 14.4 megawatts, which equates to an annualized run rate of 5.2 gigawatts. In less than 1 year, the T1 operations team has brought G1 from the start of production to a daily run rate that exceeds nameplate capacity, which speaks to the talent and dedication of our people. During the third quarter, T1 generated record net sales of about $210 million, and we expect sales to continue growing meaningfully in the fourth quarter as we start deliveries of previously booked merchant sales and we liquidate finished goods inventory that is eligible for 45X credits before year-end. This near-term sales pipeline and our continued operational progress underpin our unchanged 2025 EBITDA guidance of $25 million to $50 million. As we look forward to 2026, our supply chain team is focused on sourcing non-FEOC cells to G1 during the bridge year to the anticipated start of production at G2 in Q4 2026. We have already identified a meaningful supply of these cells for next year, which will be the primary driver of G1 production and sales before G2 is up and running. And now I'll turn the call over to Evan to walk you through the financials.

Thank you, Dan. Let's move to Slide 10 for a summary of our unchanged guidance. As detailed in this morning's release, our 2025 EBITDA guidance of $25 million to $50 million based on a 2025 production of 2.6 to 3 gigawatts is unchanged. In the fourth quarter, we anticipate a significant ramp in production and sales related to higher production levels, delivery of previously booked merchant sales as well as some liquidation of finished goods inventory before year-end. We expect fourth quarter production and module sales to exceed combined production and sales in the first 3 quarters of 2025 as we've now ramped the facility to average 4.5 gigawatt run rate in the fourth quarter. In our October release of preliminary third quarter results, we also introduced annual run rate EBITDA guidance of $375 million to $450 million for an integrated production of G1_Dallas with the first 2.1 gigawatt phase G2_Austin. The guidance is based upon G2_Austin achieving full run rate production and sales of 2.1 gigawatt and an annualized G1_Dallas run rate production sales of 5 gigawatts, supplied by 2.1 gigawatts of G2 cell and the remainder through a combination of non-FEOC foreign cells. Any U.S. sales procured potentially through Talon represents upside. Now let's turn to Slide 11 for a summary of T1's financial condition. Bringing the first phase of G2_Austin online to deliver a step change in T1's profitability and cash flow generation. The recent capital markets transactions Dan highlighted have advanced that future. Even prior to the equity transactions, our cash position built significantly as we anticipated in the third quarter. We ended 3Q with cash, cash equivalents and restricted cash of $87 million, $34 million of which was unrestricted. We added $118 million of cash in October. In addition, we accrued $93 million of Section 45X production tax credits through 3Q, and we expect to monetize those credits in the fourth quarter. We are currently exchanging term sheets. Aligned with our 4Q production and sales ramp, we expect to generate a similar amount of 45X credits in the fourth quarter that we expect to monetize in 1Q '26. On capital formation, we're building on the momentum of the recent equity transaction with potential G2 offtake contracts and debt investors. We also expect the recent equity raises will yield additional benefits for T1 shareholders. Our improvement in our capital -- our market capitalization and daily trading volume should further expand T1's eligibility for inclusion in passively managed index funds, and we are receiving a noticeable increase in inbound inquiries from active managed institutional funds who were previously unable to invest due to our trading and liquidity constraints. Now I'll turn the call back to Dan for closing remarks.

Thanks, Evan. Turning to Slide 12. Let's conclude with an overview of T1's top priorities. In the near term, our focus is on preserving T1's eligibility for Section 45X credits by completing the de-FEOCing process as well as raising the capital required to complete the first phase of G2_Austin through a combination of debt and cash deposits tied to anticipated customer offtake contracts. While we advance our capital formation and count down to compliance initiatives, we're also executing our plan for 2026, which we view as the bridge year to establish an end-to-end U.S. PV solar supply chain. Our top operational priority for the next year is to source a meaningful supply of non-FEOC solar cells to feed module production at G1 prior to the expected start of operations at G2 in Q4 2026. Concurrently, as we build the G2_Austin offtake portfolio, we intend to initiate and complete the capital formation initiatives required to fund and trigger the start of construction for the planned second phase of G2 sometime in 2026. In 2027 and beyond, we will be focusing on bringing T1's integrated U.S. supply chain online and completing the second phase of G2. We plan to achieve 5 gigawatts of integrated production between G1 and G2. And by virtue of our supply agreements with Hemlock/Corning and Nextpower, we should be producing modules of domestic content that comfortably qualifies our offtake customers for ITC stacking bonuses. Our ultimate objective at T1 is to generate shareholder value by establishing a differentiated competitive position as the first fully integrated U.S. polysilicon-based solar module producer. As we grow our operations and commercial enterprise, we will work to maximize returns on capital, sustainably reduce unit cost of production through software and automation upgrades and optimize T1's balance sheet. This is an exciting time for T1, our investors, employees, customers and partners. We are building something that doesn't exist in the U.S. today, an integrated secure, traceable polysilicon-based supply chain based on advanced solar technology. On behalf of T1's Board of Directors, thank you for your continued support in this journey as we position T1 to power America. And with that, I'll turn it back to Jeff to coordinate Q&A.

Thanks, Dan. Shannon, I think we're ready to open the line for questions.

[Operator Instructions] Our first question comes from the line of Philip Shen with ROTH Capital Partners.

Congrats on all the progress you're making. Yes, I wanted to check in with you guys on your de-FEOCing process to see if you guys could give us more color on the progress you've made and the main next steps that you guys have to take that we can follow to monitor that progress.

Thanks Philip for that. We actually have Andy Monroe, who is our Chief Legal and Policy Officer on the line. Andy, why don't you take that question?

Sure. Thanks, Dan and Philip. We're well positioned for compliance with our domestic and non-FEOC supply chain plans. We have a solid compliance plan developed with the assistance of world-class legal and compliance experts and we're making real progress on executing that plan. So we're confident. We're not sharing full details on the compliance for competitive reasons at this point, but we are confident that with those factors in play that we will be compliant.

Okay. And then as it relates to the Q3 contract dispute, could you give us a little bit more context there? Could that dispute extend longer? What kind of impacts could that be? And then how big of a contract was it? It seems like with the impairment of $50-ish-plus million, it was quite meaningful.

Yes. Thanks, Phil. Evan, why don't you take that? And as it relates to the size of the contract, we are limited to certain confidentiality on the contract. And as you can appreciate, if we are in negotiations -- or as we are in negotiations there, we have to be sensitive to the confidentiality required in the contract. Evan, would you like to add other parts?

Yes. I mean I would say that we had already calculated that in our guidance. So there isn't necessarily a guidance change as it relates to this contract, and we are continuing to execute other contracts. So in terms of the financial effect, it's been in our guidance for 2 quarters now. There was goodwill because it relates to a contract that was executed when we made the acquisition. That's why there's recording a goodwill, which we made a conservative interpretation to write off that goodwill. But as Dan mentioned, we remain in discussions with the contract party. We continue to assess all options, and we'll choose a path that optimizes the value to shareholders. I hope that's helpful.

Okay. And then one more here. You guys have made some interesting and useful -- well, interesting investments and partnerships with Nextpower and Talon here. So I was wondering if you might be able to describe more the integration of all these companies and relationships. So specifically Nextpower, what's the volume timing? When could initial modules with U.S. frames come off your line? And then as with Talon, would you expect to source cells from them to support your G1 facility? And then finally, if there's an update with Corning and Hemlock, that would be great as well.

Thanks, Phil. We are very committed to both an integrated -- vertically integrated supply chain and solar industry. So a lot of these projects are related to that. The second part of this is that domestic content. Frames are an increasingly large part. And as we go into the future, there will be a higher requirement for domestic content. A lot of the strategy around Nextpower was meeting that domestic content. As you know, beyond cells, we're basically looking at glass, at frames, at glues, at J-Boxes, et cetera. So this, to us, was a very strategic step to partner with a great company like Nextpower. I think also the Nextpower aspect was about scaling. Nextpower is a very confident partner in their products and how they scale. And we felt that having a partnership with Nextpower for these steel frames allows for the expression of that scaling from Nextpower that we could benefit through having a better customer experience from our modules. So that was another dimension of this beyond just the quality of that. In terms of volumes and timings of that, we'd expect to use that increasingly over into, if not '26 into '27, but we haven't disclosed the volumes there. Those are confidential under the contract. So we prefer to -- we'll make future disclosures on the volumes we're doing for Nextpower. As it relates to Talon , Talon was an opportunity to invest a small quantum, not disclosed in a minority position, where it would allow us to begin to talk to and look at and work with Talon in more detail. Talon is looking to build TOPCon cells. And yes, there is a way for us to procure those cells in the future. And to the degree, we have mixtures of different options in terms of cell supply, we could sell the cells to third parties, also many different options. But we're trying to reinforce and build around us the domestic chain that we really believe in. Last part on Hemlock and Corning -- that, as we've disclosed, we have optionality to convert our polysilicon to wafers. We're excited about those wafers to come from Michigan right into our G2 facility. I would comment too that our G2_Austin facility is discrete from Talon. These are 2 different projects. We're excited about our project, and we're excited about our minority investment in Talon.

Great, Dan. Looking forward to seeing the full results of your integrated supply chain.

Thanks.

One more, if I may. This is from an investor. He's asking how is T1 claiming or planning to claim the 45X credits in terms of stacking when they produce cells in one site and modules at another site when the OBBA says they have to be at the same facility?

Andy, do you want to take that, please?

Sure. Without getting into all the details, there are provisions in the Act that allow for the election of unrelated party transactions, and those provisions have not been changed. That was in the original Act and were not changed by the OBBA.

Our next question comes from the line of Greg Lewis with BTIG.

Guys, I was hoping to get an update on kind of how we should be thinking about the event path for G2. Any kind of hurdle rates we should be thinking about in the next couple of quarters, just as we think about getting that facility up and running by the end of '26 to really set the table for '27 production.

Thanks, Greg. I'll have Otto layer in here, too. We've been working very hard for the last year to design the right paths here. We have over a 30% design done. We have work packages out that are live. As you know, we did raise capital earlier this last month -- this month to unlock some capital in order to begin the first stages of construction. We are still on track to go and start production -- I'm sorry, start construction in the fourth quarter of this year. The paths really go to the site, the equipment, the machines, the early earthworks and concrete and steels packages. Those are the biggest time lines in terms of risks to the time line. As Otto mentioned in his remarks, the steel package was particularly important and some of the switchgear was particularly important. Beyond that, if we look at the equipment, the equipment is not on a critical path, but we wanted to advance those work packages and get those equipment orders as fast as possible also. Otto, do you want to talk about the cadence and how we're tracking toward the fourth quarter '26?

Sure, Dan. So yes, so as you mentioned, really, it's all about getting started now, getting started with earthworks, preparing to erect steel in March and also securing the long lead items. So electrical equipment, we've talked about as well, there's air units, there's other utility systems like water and utility plants that needs to come in place. So it's all about getting started and execute those contracts that we have lined up and are negotiating now as soon as possible. So we're tracking towards our time line.

Okay. Great. And then just I wanted to go back to Slide 6, where you kind of outlined the -- clearly, what's going on in power -- power is cool again, right? And so as we think about that and kind of the acceleration and the potential for solar, if you go back and look, like no one -- I feel like no one's really -- you don't hear data centers talking about solar. I mean, last year, we installed 50 gigs in the U.S. I think it was a few gigs of natural gas. And just -- so as we look at meeting this increasing demand for power gen in the U.S. are we getting the sense that we hear a lot about behind the meter are hyperscalers pursuing this or other entities? Or do you think really the bulk of this solar growth that we're going to see in the U.S. over the next 5 to 10 years. Is that largely just going to still be with utilities?

Yes. We're seeing tremendous interest from developers and it's a pass-through basically data centers, AI companies. The utility scale levels and the quantum of power that's needed, it's really only the things that solar, which we do and storage together are only thing that's going to deliver that until basically 2029, 2030 when natural gas gen hits or nuclear starts coming back. We fully believe in a combined industry that is supportive of multiple uses of energy and all of the above strategy, but solar is the only thing that's scalable right now. When we -- when the U.S. looks -- when you look at China, China has over a terawatt of manufacturing capacity across ingot wafer cells and modules, a terawatt of manufacturing capacity. First half of the year, China put in 256 gigawatts. So there is tremendous human intelligence and tremendous scope to really deploy it. And we do think that the United States has those elements of capital, has those elements of technology to start building that, and we'd like to see more of that develop in the U.S. But solar is the answer right now. I do think we've reached the tipping point in terms of the costs, in terms of particularly the storage costs and the adjacency to solar. And I think those 2 things are delivering. I do think that building these projects and designing them with either natural gas in mind or other longer-term grid access in mind is an important dimension. And the last part I'd say is I think a lot of these other places are really going to be about distributed energy resources, energy islands. The amount of power that AI needs and the ramp that AI wants, it's just too hard to do that at current grid and current connections. So we're very confident on the future of how solar is going to contribute into that energy.

Our next question comes from the line of Sean Milligan with Needham & Company.

Just a quick question. It looks like you mentioned that you've ramped up G1 now to over 5 gigawatts. I'm curious about how you see that sustaining into 2026. And then what you're seeing for demand in 2026 there? And then just looking forward, kind of what you're seeing for demand in 2027 as G2 comes online? And kind of the third part of that question is another publicly traded company made some comments about pricing on their call. So is there any kind of like pricing guardrails you can give us for kind of non-FEOC cells in '26, what you're looking at and then also 2027 with G2 online?

Yes. Thanks for that. What we've seen in this year is that we've had a very, let's say, erratic market in solar with -- is the OBBB going to kill the IRA. It did not. You have demand looking at this 232 coming, what is the teeth it's going to be. So the industry has been dealing with inventory, a lot of sales uncertainty. This uncertainty has made for a very choppy 2026. I think that ties to a lot of how we have a back-end loaded volume in 2025. So that really explains the landscape of what we've had today. As we look into 2026, which is a bridge year for us, we will not expect to produce and we will not produce domestic cells. Those are expected to start coming on in the fourth quarter of 2026. So as those come out in the fourth quarter 2026, that will be towards -- that will only be part of it. But for '26, we have to source non-FEOC cells. We feel confident that we have the ability to source quantums, but we are not yet coming out with our guidance there in terms of what we'd like to express. On pricing, it's complicated also because the pricing of those non-FEOC cells is also a question. So we'll be looking to come out with guidance for 2026 and give that pricing update and those volume updates for '26. When I look at '27, which is what we're very, very focused on, which is the domestic cell, that's where we're in active discussions with large utility scale type investors. And we do see demand. We do see strong interest there. There is strong interest in the domestic selling, domestic module, and that's what our focus is. As we get those offtake discussions or contracts done, we will, of course, be disclosing those in full. But the focus really is about how to start delivering in '27. Evan, do you add anything color to the pricing or to the volumes?

Yes. No, no. Look, I think you covered it, Dan. I mean, look, demand is high, right, for '26. It's going to break it up, right? And we're seeing early prices that are higher than current pricing, right? So several cents a watt higher than where we are currently in the fourth quarter. It's going to be cell availability that drives production levels more so than demand. As Dan mentioned, we've begun -- we have attractively priced non-FEOC cells in our inventory today, and we are working aggressively to procure those for 2026, which is our bridge year. But I think that's what's going to drive your value. And we'll provide production range here shortly. For 2027, that's where -- at least for Phase 1, right, Phase 1 of G2, you are in a lot of conversations with parties that have a demand that far exceeds our 2.1 gigawatt production, right? So -- and those discussions are for multiyear offtake contracts that are very attractive, okay? And so we expect to, over time, certainly by the time we're producing a facility to have most, if not all, of that volume contracted the 2.1 gigawatt. And then it becomes a question of how quickly can we convert excess demand for G1 into -- sorry, for G2 Phase 1 into an underpinning for G2 Phase 2, right, which, again, we think it's going to be driven by offtake demand, but we clearly see the potential for that following in some reasonable or short time period from financing on G2 Phase 1, right? The goal would be ultimately to put as many of the high-margin in-demand cells into G1 as possible as quickly as possible. I don't know if that gets...

That's great. That's great. And then the other question was on the COGS side. So this year, I know you've been doing a lot with your supply chain. And then next year, you bring on non-FEOC cells. I'm just curious how you see COGS moving around this year and if that starts to normalize some next year as you kind of get up to scale more?

Look, that's a good observation. I think you'll see it in the fourth quarter, right? Obviously, when you're at scale at a level that's averaging 4.5 gigawatt run rate in the fourth quarter, your conversion costs come down significantly throughout the course of the year, and we see a forward path to a facility in its second year of operation to continue to make gains on those costs that we control. As it relates to procurement and pricing, again, we are seeing -- your cell is most of your cost, but throughout the [ BAM ] we continue to work to optimize that, and we expect to make improvements. Again, we were ramping a facility into a period that had unusual tariff volatility. So it was like you were less able to kind of optimize timing of costs and you were in a period where rising tariffs, you were hit by some of those tariffs. We think a lot of those risks will be mitigated even in an environment where 232 impacts the market, given we have a differentiated and advantaged supply chain. So we'll provide further quantification of like some of those improvements when we, in near term, put out our 2023 guidance, which we're again making traction on locking things in.

Yes. I would just add to Evan that you touched on the polysilicon side. As you know, the cell is the bulk of the cost, and we work diligently to ensure very competitive cells. Our company, all of our polysilicon is from Hemlock. We take the polysilicon from Hemlock that's turned into wafers in Vietnam. We have control of the polysilicon side. And the reason I mentioned this is with the anticipation of what may come out of 232, we feel that we're very protected on that cost element. Again, we get the benefit of basically having a locked-in pricing on our polysilicon. So to the degree 232 does come out and does add cost to other non-American polysilicon or Chinese polysilicon, we think that we're in a very advantaged state as that feeds through into the cell costs.

Great. That's great, Dan. And then on Section, the 45X tax credits. I know this year, you've built up a good amount on the balance sheet, and you said you're looking to monetize those currently. It's not like swapping term sheets. As we look forward, should we think about credit monetization being a more regular step in the process for you all? Or is it going to be kind of larger transactions single time like once a year? Or are we thinking multiyear type transactions there to help with liquidity?

I think you're spot on the cadence, I'm going to let Evan cover some of the details. We came -- started fully commissioning full certificate of occupancy in the first half. We did get all of our first half volumes in terms of what was produced, and then we've been out in the market doing that right into the face of OBBB. So there was a lot of uncertainty around the world about those aspects. So I do expect on a go-forward basis, there will be a much more normal cadence on how we monetize 45X. And then on the other side of 45X direct pay versus selling through banks to third parties, that also is an element that we wanted to make sure we optimize in terms of the prices and the costs that we are trying to get there. Evan, do you want to talk about the timing of when we would expect to see 45X now?

Yes. Look, I mean, I think as I said in my comments, we expect to execute third-party sales in this quarter for all or almost all of the 45X that we generated in 2025. I think on a go-forward basis, yes, we're looking to enter into a quarterly cash settle within some number of days after the quarter with one or several parties for our volumes. I think '26 is -- it's a year that has newer requirements that are different from the past. So it might be a slower to develop year. So I think they will be more midpoint of the year and on. But kind of going forward, I think it will be more traditional of, again, quarterly cash settle on a third-party sale, right, versus direct pay.

All right. Congratulations on the continued move forward.

Thanks, Sean.

This concludes the question-and-answer session. I would now like to turn the call back over to Jeffrey Spittel for closing remarks.

Thank you, Shannon. Thanks, everybody, for the interest. We will be back on the road at conferences in New York next week. Please feel free to reach out with additional questions, and thanks for the interest and participation today. This will conclude the call.