Operator
Good afternoon, and welcome to the NETLIST 3rd Quarter 2026 Earnings Conference Call and Webcast. All participants will be in a listen-only mode. Should you need assistance, please signal a conference specialist by pressing the star key followed by zero. After today's presentation, there will be an opportunity to ask questions. To ask a question, you may press star, then one on a touch-tone phone. To withdraw your question, please press star, and then two. Please note this event is being recorded. I would now like to send the conference over to Mike Smarjassi, Investor Relations. Please go ahead. Thank you, Nick, and good day, everyone.
Welcome to Netlist's third quarter 2026 conference call. Leading today's call would be Chuck Hong, Chief Executive Officer of Netlist, and Gail Sasaki, Chief Financial Officer. This quarter, we have prepared presentation slides to accompany management's remarks and provide additional context around our financial results and recent developments. The slides, along with today's earnings release, are available in the Investor Relations section on NetList's website at netlist.com. A replay of today's call will also be available on the website following its conclusion. Before we begin, I would like to remind everyone that today's remarks and responses to questions may include forward-looking statements based on current expectations. Actual results may differ materially from those expressed or implied in these statements due to risks and uncertainties outlined in today's earnings release and netlist filings with the SEC. Netlist undertakes no obligation to update any forward-looking statements. With that, I will now turn the call over to Gail.
Thanks, Mike, and good afternoon, everyone. I'll review our third quarter financial results, discuss the key drivers of our performance, and then provide some perspective on our financial position and outlook. For those who want to follow along, we will start with slide three. Net sales for the third quarter of 2026 of $105.1 million grew 149% from $42.2 million in the prior year period. License fee revenue of $44.9 million was recognized in the quarter, driven by revenue recognition under our Samsung agreement. Samsung's upfront gross license fee of $239 million is being recognized as revenue rateably over the 20-quarter term of the agreement, which equates to approximately $12 million per quarter. Licensing revenue recognized in the third quarter included this amount, plus Samsung quarterly license fee of $32.9 million. Net product sales of $60.2 million increased 43% compared to prior year period, driven by strong demand from customers in support of AI and data center applications. Total gross margin rose to 65.5%, an increase of more than 61 percentage points from 4.4% in the prior year period. While product growth margin improved to 40% compared to 4% in the third quarter of 2025. The year-over-year increase was primarily led by the growth in licensing revenue, driven by our Samsung agreement. Our focus on netless product opportunities, specifically the Lightning Ultra Low Latency Memory product line, and the strong overall market environment. Total operating expenses for the quarter were $28.3 million, up from $9 million in the prior year period, reflecting higher intellectual property legal fees and other non-recurring expenses related to the Samsung and Micron transactions. Intellectual property legal fees rose to $16.2 million from $5 million in the prior year quarter, again, primarily in connection with our Samsung and Micron litigation and settlements. Research and development expenses increased to $1.1 million from $0.8 million in the third quarter of 2025, primarily due to higher employee headcount and the associated increase in overhead costs. Selling general and administrative expense was $11 million, up from $3.1 million a year ago. The increase was primarily due to higher stock-based compensation expense, sales commissions associated with higher sales, And non-recurring expenses related to the Samsung and Micron transactions. Looking ahead, we continue to make investments in sales, marketing, and R&D to support growth in our product business, expand our customer base, and capitalize on opportunities under our Samsung supply agreement. At the same time, we anticipate intellectual property legal expenses will moderate to levels more consistent with the first quarter of the year. Operating income for the quarter was $40.6 million, compared with an operating loss of $7.2 million in the prior year period. Income before income taxes was $41.5 million. During the quarter, we recorded an income tax benefit of $44 million, primarily related to the release of the valuation allowance on deferred tax assets, including our net operating loss carry forwards. This adjustment reflects the company's recent performance and our improved outlook for future taxable income, as well as our resulting assessment that it is now more likely than not that we will be able to realize these tax benefits. The release of the valuation allowance favorably impacted our reported effective tax rate and net income for the quarter. Net income was $85.4 million, or $0.22 per diluted share, compared with net loss of $0.7 million, or a loss of $0.02 per share in the third quarter of 2025. Turning to our balance sheet and cash flow, we ended the quarter with $327.7 million in cash, cash equivalents, and restricted cash. Up from $40.7 million at June 27, 2026, primarily due to Samsung's initial cash payment of $227.5 million, which includes the upfront license fee of $200 million net of taxes and the first quarterly payment of $27.5 million net of taxes. We continue to have no debt outstanding. Cash flow from operating activities was $245 million for the quarter, and was largely driven by the deferred revenue related to the receipt of the Samsung payments that I just mentioned. We're now on slide four, and I'll talk about the Samsung alliance. I'd now like to quickly summarize our agreement with Samsung. As we announced previously, we signed a five-year strategic alliance with Samsung in August. Under which Samsung licensed our entire patent portfolio. We also secured a product supply agreement under which Netlis has the right to purchase up to $300 million per year of Samsung DRAM and NAND at a competitive cost. Under our licensing agreement, we expect to recognize $898 million in license revenue over the five-year term. Which consists of the $239 million gross upfront license fee, which will be recognized over the 20 quarters of the agreement, or approximately $12 million per quarter, and a recurring quarterly gross license fee of $32.9 million. Slide 5 is the Micron Agreement. Turning to our Micron Agreement, which we signed in early October. Under the agreement, Micron licensed our entire patent portfolio and purchased 10 million shares of Netlis stock. In exchange, Netlis will receive $30 million in quarterly license fees starting in the fourth quarter of 2026 through the third quarter of 2031 for a total cash license value of $600 million over the term of agreement. Quarterly revenue recognized under this agreement will be reduced by approximately $2 million per quarter of non-cash expense, reflecting the value of netless stock acquired by Micron. Beginning in the fourth quarter, we will recognize license fee revenue under both our Samsung and Micron agreements. We expect a combined quarterly licensing revenue of $73 million, $292 million per year, or $1.46 billion of recurring revenue through the end of the five-year license terms. Our Q426 outlook is on slide 7. We expect Q426 operating income to come in substantially higher from the third quarter with the addition of Micron's recurring license payment. We expect product sales to come in substantially higher driven by growing demand for netlist products and the initial sales of Samsung products under our product supply agreement. We will continue to manage our operating expenses, including the variability associated with intellectual property legal activities, while maintaining appropriate investment in our product roadmap. I will now turn the call over to Chuck for an overview of our strategy and near-term priorities.
Thank you, Gail, and good afternoon. We continue to execute against our strategy of building Netlis's position in the AI memory landscape while developing and protecting the intellectual property that underpins our technologies. Today, memory is a major bottleneck in the AI infrastructure. With a massive build-out of AI data centers, there's an insatiable demand for high-performance memory used for both training and inference. That demand, presently, cannot be filled by global supply. With that as backdrop, our recent agreements with Samsung and Micron, the world's number one and number three memory manufacturers, serve as a reminder of the critical importance of Netlis's intellectual property in the AI ecosystem. These licenses were among the largest dollar value in the history of the memory industry. The agreements provide a strong financial and operational foundation for Netlis' future growth. The recurring royalty stream complements the company's existing products business while providing long-term revenue visibility. The alliance with Samsung creates opportunities for technical collaboration on advanced memory products. And provide supply of scarce DRAM and NAND raw material at a competitive cost in a short supply environment. The landmark agreements also provide groundwork for building a sustainable, repeatable approach to monetizing our IP portfolio and establish industry benchmarks for the value of netless IP for other licensees as well as for future renewals. We will continue to pursue our intellectual property strategy in a disciplined manner, including protecting our patents, pursuing additional licenses, and evaluating opportunities to expand the reach and value of our technology. Our focus now is to translate these added resources into product development, customer execution, and disciplined growth. Netlis has spent years developing technologies and building an IP and products portfolio in advanced memory, including server DEMs, HBM, and enterprise SSDs. We believe our combination of product capabilities, engineering expertise, and patented technologies gives us a unique platform to convert opportunities to tangible business across AI segments and related markets. As we look ahead, our priorities remain focused on execution. First, we will accelerate the development of our next-generation AI memory technologies and pursue opportunities in segments where high-performance memory are increasingly important, such as hybrid CXL for KVCache. Second, we will prioritize discipline execution. Including managing working capital, maintaining appropriate cost controls, and improving the sustainability of our financial results. We plan to complement this organic growth with select acquisitions to expand our opportunities in the edge AI space. In closing, we plan to build on Netlis's advanced memory expertise, strengthen our operating performance, and fully realize the potential value of our intellectual Thank you. Delivering on that opportunity will require continued investment, execution, and a clear focus on creating value for our shareholders. On behalf of the entire Netlist team, thank you to our customers, partners, employees, and shareholders for your continued support.
Thank you, Chuck. Nick, please open the call for questions.
Operator
Thank you. We will now begin the question and answer session. To ask a question, you may press star, then 1, on your touchstone phone. If you are using a speakerphone, please pick up your handset before pressing the keys. If at any time your question has been addressed and you would like to withdraw your question, please press star, then 2. The first question will come from Suji DeSilva with Roth Capital. Please go ahead.
Hi, Chuck. Hi, Gail. Congratulations on all the progress here. Maybe, Chuck, we can start with the product business and the Lightning product. I'm just trying to understand, you know, the opportunity there for agentic AI and inference. Is it, you know, really the inference KV cache growing size, or is it the CPU intensity of the agent jobs? What's the biggest? You know, driver in the AI inference area that requires Lightning, and, you know, where are the substitutes that would compete with Lightning that it has an advantage over, and what's that advantage? Thanks.
Yeah, Suji. So, the ultra-low latency memory modules are an optimal solution to the... To the proliferation of KVCash. So KVCash is pretty much the storage of tokens. As we put inquiries, as the AI models get built, tokens are generated, and those tokens have to be stored otherwise. If our inquiries or queries are not kept in storage, you know, they'll have to restart, and that will be very costly. So KVCash is probably one of the best applications for ultra-low latency, although we started this product line to be focused more on the high-frequency trading. Where milliseconds, microseconds are also very important, high performance. So it's been a very good product thus far. We continue to expand the product line, and I think the KVCash for AI probably is an application that's going to continue to help this product line grow.
Excellent. Now, looking forward to the product progress there. And then switching over to the litigation side, licensing side, any framework or sequence discussion on the additional potential licensing activity, maybe even sizing now that we have Samsung and Micron as reference points, you know, the sequencing that would optimize positive outcome for you guys, any color there would be helpful.
So, we have Hynix and Micron, and there is the third largest, they're the one in three largest manufacturers of memory. Samsung and Micron, sorry, and then you have Hynix. So, I think we can deduce what the, you know, the next. Logical license would be. There are other significant suppliers of memory that are not in the major three, and we will continue to pursue other significant parties outside of the three major memory suppliers. I think the important thing is that the two Samsung and the Micron licenses establish a very strong benchmark for the value. So it's a comp for the value of our intellectual property. So we will continue to work on other licenses as well as renewals as well.
Great. And then one last question. It looks like the product area is now a significant focus for you. Can you just kind of prioritize between organic product introductions, acquisitions, and potentially the Samsung and Micron, or more Samsung collaboration? And if you expect to have fruits from there in terms of kind of co-developed products, which of the three avenues may be the more likely way that you'll grow your product offerings?
Well, with the limited resources that we've had, we've not been able to put as much engineering resources into the existing projects that we talked about, such as the Hybrid CXL. Hybrid CXL, which was a storage class memory application initially, it is now... We believe would be a perfect product for, again, for KV cache storage. A lot of KV cache is stored first on HBM. It flows over into DEMS. Now a lot of this is being stored on SSDs. And so our Hybrid CXL sits. At a tier above SSDs in terms of performance. CXL is much faster than NVMe SSDs. So we are going to accelerate that project by bringing in and recruiting as many design engineers as we can in a short period of time. We will continue to work on, also do the same for low-power MR-DEMS, which is also another KV cache tier. But over time, we believe that MR-DIMS will displace our DMS as the central platform for server memory. So those two projects will double down in terms of our R&D. Resources in those areas. It's certainly past the research stage. It's in full development, but those two areas, we will also be looking at. We are looking at the edge AI space. The MR-DEM and the TXL products is AI infrastructure. You can put that in infrastructure, main compute, kind of cloud computing, data center space. And then there is the edge AI, which is AI processing. That is disconnected from the cloud. And those are for applications in the military, in industrial, ruggedized, medical, so forth. And we are looking at types of computational storage technologies. And we've got some of that in-house. We're looking for some outside technology that could complement what we have. So that would be a separate space. If we look at infrastructure and edge as two distinct, you know, AI spaces, edge is going to require More compressed, smaller AI models that can be detached from the data center. So our ongoing R&D projects have been in the infrastructure space. We are now looking for expansion into the edge AI space.
Excellent. Thanks, Chuck. Thanks, Gail. Good to see progress in many fronts here. Thanks.
Thanks, Suji. Thank you, Suji.
Operator
This concludes our question-and-answer session as well as conference call. Thank you for attending today's presentation. You may now disconnect.
