Arteris - Earnings Call - Q1 2025

Executive Summary

• Q1 2025 revenue was $16.5M, up 28% y/y, exceeding the top end of guidance and aided by ~$0.5M one-time revenue; non-GAAP EPS was $(0.09), non-GAAP gross margin 92%.
• Record ACV + royalties of $66.8M and RPO of $88.9M underscore demand momentum across enterprise computing, communications, and automotive; free cash flow was +$2.7M.
• Management widened FY25 guidance ranges (revenue $65–$71M; ACV exit $71–$79M; non-GAAP op loss $7–$14M; FCF $0–$8M) due to tariff/geopolitical uncertainty and FX headwinds (approx. $1M expense impact if USD weakness persists), while keeping midpoints broadly unchanged.
• Near-term stock narrative catalysts: accelerating AI and chiplet adoption (FlexGen in >20 evaluations, expected bookings 2H), record ACV/RPO, Intel Foundry Accelerator IP & Chiplet Alliance participation, and potential customer outsourcing tailwinds.

What Went Well and What Went Wrong

What Went Well

• Record ACV + royalties of $66.8M and RPO of $88.9M; “we delivered record annual contract value plus royalties of $66.8 million and generated $2.7 million in positive free cash flow”.
• Strength in AI and large accounts: “4 came from top 30 global technology companies… another was an expanded reorder from a top 5 technology company” and AI-related deals now account for “north of 55%” of business.
• FlexGen traction: >20 SoC projects evaluating; production status reached in Feb (FlexGen 1.2), expected to drive bookings and revenue ratably starting 2H 2025.

What Went Wrong

• Macro/tariff uncertainty and FX headwinds: widened FY guidance ranges; ~40% of expenses in foreign currencies and euro appreciation could impact expenses by ~$1M if sustained.
• Non-GAAP operating expense up 9% q/q and 8% y/y (investment in R&D, FAE, sales; weaker USD effect), with GAAP operating loss of $(7.7)M.
• Potential short-term royalty headwinds amid tariff/geopolitical backdrop and confidence softness, though licensing remains robust; some project replanning in China.

Transcript

Operator (participant)

Good afternoon, everyone, and welcome to the Arteris Quarter 2025 Earnings Call. Please note this call is being recorded and simultaneously webcast. All material contained in the webcast is sole property and copyright of Arteris, with all rights reserved. For opening remarks and introductions, I will now turn the call over to Erica Mannion of Sapphire Investor Relations. Please go ahead.

Erica Mannion (Partner and Founder)

Thank you, and good afternoon. With me today from Arteris are Charlie Janac, Chief Executive Officer, and Nick Hawkins, Chief Financial Officer. Charlie will begin with a brief review of the business highlights for the first quarter ended March 31, 2025. Nick will review the financial results for the first quarter, followed by the company's outlook for the second quarter and the full year of 2025. We will then open the call for questions. Before we begin, I'd like to remind you that management will make statements during this call that are forward-looking statements within the meaning of federal securities laws. These statements involve material risks and uncertainties that could cause actual results or events to materially differ from those anticipated, and you should not place undue reliance on forward-looking statements.

Additional information regarding these risks, uncertainties, and factors that could cause results to differ appears in the press release Arteris issued today and in the documents and reports filed by Arteris from time to time with the Securities and Exchange Commission. Please note, during this call, we will cite certain non-GAAP measures, including non-GAAP net loss, non-GAAP net loss per share, and free cash flow, which are not measures prepared in accordance with U.S. GAAP. The non-GAAP measures are presented as we believe that they provide investors with a means of evaluating and understanding how the company's management evaluates the company's operating performance. These non-GAAP measures should not be considered in isolation from or as substitutes for or superior to financial measures prepared in accordance with U.S. GAAP.

A reconciliation of these non-GAAP measures to the nearest GAAP measure can be found in the press release for the quarter ended March 31, 2025. In addition, for a definition of certain of the key performance indicators used in this presentation, such as annual contract value, contract design starts, and remaining performance obligations, please see the press release for the quarter ended March 31, 2025. Listeners who do not have a copy of the press release for the quarter ended March 31, 2025, may obtain a copy by visiting the Investor Relations section of the company's website. In addition, management will be referring to the Q1 2025 earnings presentation, which can be found in the Investor Relations section of the company's website under the Events and Presentations tab. Now, I will turn the call over to Charlie.

Charlie Janac (CEO)

Thank you, Erica, and thanks to everyone for joining us on our call today. In the first quarter of 2025, we achieved another record annual contract value plus royalties of $66.8 million and generated $2.7 million in non-GAAP positive free cash flow as demand for commercial semiconductor system IP products continues to grow. Our success during the quarter saw steady adoption across enterprise computing, communications, and automotive semiconductors, driven by growing chiplet and SoC design complexity, as well as proliferation of AI applications. During the quarter, we had several key design wins. Four came from top 30 global technology companies expanding their deployment of Arteris products. The largest win included Magillem SoC integration automation software, as well as interconnect IP for various applications, including memory controllers and consumer electronic projects.

Another of the large wins came from an expanded reorder from a top five technology company with products and services, including hyperscale computing and consumer electronics. Also, a major automotive OEM expanded its use of Arteris product portfolio for its next generation of EV vehicles. Another key win, and one of the three new Arteris customers in the quarter, was with an industry-leading Japanese automotive OEM. This customer licensed our products to support their new in-house development of autonomous driving SoCs that include AI and functional safety capabilities. They selected Arteris based on the combination of our product's superior performance, lower power area efficiency, and high resilience for their mission-critical applications. With this latest addition, we now have 10 automotive OEMs as direct Arteris customers. Adoption of our technology also continues to be strong with advanced semiconductor companies.

For example, our physically aware FlexNoC IP with AI and functional safety support was chosen for the development of next chips, next generation vision-based ADAS technology to realize the future of autonomous driving with sustainability. We are seeing increased movement from internal system IP solutions to commercial vendors such as Arteris as customers desire resource efficiency, quality, and faster solution delivery. I'm pleased to report that our penetration of the increasingly complex microcontroller MCU system IP market continues with initial receipt of royalties from a top five MCU manufacturer. This penetration is driven by continued increases in MCU complexity, as well as ever more stringent latency and cost requirements. In addition to our customer momentum, we continue to deliver new technology.

In the last earning call, we announced FlexGen, our AI-driven smart NoC IP technology, which has the potential to revolutionize semiconductor designs by delivering up to 10x engineering productivity, lowering power consumption, and improving overall performance. We now have over 20 customer SoC projects evaluating FlexGen, which is a promising start for this innovative product, which we expect will generate revenue and ACV in the second half of the year. In the first quarter, Arteris also released the latest generation of Magillem register management automation software used for semiconductor hardware and software integration. This latest technology provides a single source of data for development of SoCs and chiplets by chip architects, hardware designers, firmware engineers, verification teams, and documentation teams, helping to mitigate the silicon failure risks associated with the unfortunate and quite common instances of out-of-date specifications, interpretation differences across various teams, and user errors.

The latest product improves performance and scalability to address the needs of any semiconductor design, ranging from simple IoT devices to state-of-the-art complex artificial intelligence SoCs, FPGAs, and chiplets. I'm also proud to see that our focus on innovation being recognized in three prominent categories in the 23rd Annual American Business Awards out of 3,600 nominations. These included the Gold Award for the Most Innovative Tech Company of the Year, another Gold Award for Technical innovation of the Year for our Ncore NoC IP with its support for Arm, RISC-V, x86, and mixed architectures, and the Silver Award in the product innovation category for our FlexNoC and Ncore NoC dialing technology in support of advanced AI computing at the data center and the edge.

Beyond actively driving in-house innovation, Arteris continues to expand ecosystem collaboration to provide full solutions to our customers, including leveraging our product's physical awareness to support the faster development of advanced electronics with more predictable power, performance, and area, or PPA, for SoCs and chiplets. We recently announced that Arteris joined the Intel Foundry Accelerator program, becoming members of the IP Alliance, enabling silicon designs using Intel's 18A advanced process node to collaborate on physically aware NoCs for future nodes. Additionally, Arteris also became a founding member of Intel Foundry's new Chiplet Alliance, which aims to create a robust network of ecosystem partners to ensure interoperability and to accelerate creation of a wide range of multi-die silicon applications.

Similarly, Arteris also joined the Imec-sponsored Automotive Chiplet Forum, whose goal is to share insights and ensure industry alignment and interoperability for automotive chiplet-based architectures, where our NoC products with ISO 26262 functional safety capabilities will play an increasingly important role moving forward. Lastly, we announced the opening of our new engineering and customer support center in Kraków, Poland. This new location will support the development of network-on-chip IP and SoC integration automation software for the semiconductor industry. Arteris has hundreds of customers worldwide who are supported by workforce across 11 countries. The addition of a hub in Poland will expand the company's global footprint and provide Arteris with expanded access to top engineering talent for product development, validation, and customer support.

We believe the scale and scope of our long-term opportunity remain robust and are supported by our current products and strong product pipeline of new silicon system IP technologies, as well as growing relationships with some of the largest and most advanced electronics companies in the world. Our customers continue to innovate in exciting high-growth areas such as generative AI, autonomous driving, 5G and 6G communications using Arteris products, and global support. We are diligently monitoring the current global economic uncertainty, although this did not lead to any deal cancellations or delays in the first quarter. Nevertheless, we do see greater potential for variability in financial outcomes for the year due to this economic uncertainty. The clearest impacts are potential short-term headwinds to royalties as a result of waning customer global confidence and automotive and other tariffs.

Additionally, our overseas-based OpEx is likely to increase should the recent weakness of the U.S. dollar persist or worsen. As a potential offsetting factor, we are seeing opportunities for our customer base to accelerate outsourcing of their silicon system IP needs to Arteris to accelerate their products' time to market, reduce their own costs, and increase their operating efficiencies. Nick will cover these impacts more when he discusses our guidance. With that, I'll turn it over to Nick to discuss our financial results in more detail.

Nick Hawkins (CFO)

Thank you, Charlie, and good afternoon, everyone. As I review our first quarter results today, please note I will be referring to GAAP as well as non-GAAP metrics. A reconciliation of GAAP to non-GAAP financials is included in today's earnings release, which is available on our website. Also, as a reminder, I will be referring to the Q1 2025 earnings presentation, which can be found in the Investor Relations section of the company's website under the Events and Presentations tab. We had a strong first quarter characterized by beating our guidance on all key financial measures. Turning to slide five of the presentation, total revenue for the first quarter was $16.5 million, up 28% year-over-year, benefiting from approximately $500,000 one-time revenue and exceeding the top end of our guidance range.

At the end of the first quarter, annual contract value, or ACV plus royalties, was $66.8 million, up 15% year-over-year, above the midpoint of our guidance range and a record high for the company. Remaining performance obligations, or RPO, at the end of the first quarter were $88.9 million, representing a 19% year-over-year increase, once again a new high for Arteris. Non-GAAP gross profit for the quarter was $15.3 million, representing a gross margin of 92%. GAAP gross profit for the quarter was $15.0 million, representing a gross margin of 91%. Now turning to slide six, non-GAAP operating expense in the quarter was $18.4 million, up 9% sequentially and 8% higher year-over-year, impacted by the timing of certain non-linear expenses and in part driven by the weaker U.S. dollar increasing the cost of our overseas operations.

Additionally, we continue to grow the investment in our R&D and field application engineering teams that drive technology innovations and solution support. Total GAAP operating expense for the first quarter was $22.7 million, representing a 10% year-over-year increase. As we look ahead, we plan to focus spending on strategically critical areas, in particular in key people who can help drive product development, enhance customer support through field application engineering, and expand the geographic and key account reach of our global sales team. We believe that these ongoing investments can accelerate our top-line growth. At the same time, we are driving operating leverage by controlling G&A spending, which has now remained broadly flat on a non-GAAP basis for approximately three years. Non-GAAP operating loss for the first quarter was $3.2 million, close to the top end of our guidance range.

This represents a $2.1 million or 40% improvement compared to the loss of $5.3 million in the prior year period. GAAP operating loss for the first quarter was $7.7 million compared to a loss of $9.1 million in the prior year period. Non-GAAP net loss in the quarter was $3.6 million or diluted net loss per share of $0.09 based on approximately 40.9 million weighted average diluted shares outstanding. GAAP net loss in the quarter was $8.1 million or diluted net loss per share of $0.20. Moving to slide seven and turning to the balance sheet and cash flow. We ended the quarter with $55.1 million in cash, cash equivalents, and investments, and we have no financial debt.

Free cash flow, which includes capital expenditure, was positive $2.7 million for the quarter, above the top end of our guidance, benefiting from some early customer payments that were due early in the second quarter. I would now like to turn to our outlook for the second quarter and full year 2025 and refer now to slide eight. Looking forward, the current economic turbulence has created some marked uncertainty for our business, and consequently, we have revisited our 2025 guidance. As a general contextual comment, our view is that the current economic turbulence presents three key financial considerations for 2025. First, the current trade challenges may result in short-term reduction for end demand for some of our customers' products and in our key market verticals, especially automotive and consumer. It is, however, not yet clear what, if any, impact there will be to our royalty revenue in 2025.

The first quarter unit sales outputs from our customers have generally been better than expected. However, it is possible that this is in part resulting from pull-forward demand for our customers' products in anticipation of increased tariff costs. Consequently, we have not adjusted our overall FY2025 revenue guidance at the midpoint since we believe that the overall net impact will not be materially different from our prior expectations. Second, since the start of the year, the US dollar has weakened against most major currency pairs. While the significant majority of our revenue is invoiced in US dollars, approximately 40% of our expenses are denominated in foreign currencies, predominantly the euro, which has appreciated by up to 10% against the U.S. dollar this year.

In the event that the U.S. dollar exchange rates remain at current levels, we estimate that the annual impact on our tariff expenses would be approximately $1 million. However, due to unrelated offsetting expense factors, we are not adjusting our midpoint guidance for non-GAAP operating income and free cash flow for FY2025. Third, while our tariff products are not subject to tariffs, there is a potential existential impact of the ongoing trade disputes and collateral economic impacts to our business environment, including factors such as consumer and industrial confidence. As a result, we have widened our top-line guidance ranges. This economic turbulence is exogenous to our tariff business operations, and it is hard to forecast with certainty the longevity or collateral consequences of changing our economic policies.

That being said, while the industrial markets remain clouded with tariff and geopolitical uncertainty, we see our customers' long-term growth and therefore our license and royalty revenue remaining robust. For the second quarter of 2025, we expect ACV plus royalties of $66 million-$70 million, revenue of $16.1 million-$16.5 million, with non-GAAP operating loss of $4 million-$3 million, and non-GAAP free cash flow of negative $5 million to zero, which reflects the reverse effect of early customer payments that benefited the first quarter, as I mentioned earlier. Therefore, we expect free cash flow for the first half overall to be positive at the midpoint.

For the full year 2025, our guidance is as follows: ACV plus royalties to exit 2025 at $71 million-$79 million, revenue of $65 million-$71 million, non-GAAP operating loss of between $14 million and $7 million, non-GAAP free cash flow of zero to positive $8 million. In spite of the near-term challenges I outlined, we are very encouraged by the continued strong deal pipeline, and I would reiterate the point raised earlier by Charlie that we are seeing promising signs of an accelerated interest by some major customers to increase their outsourcing to the commercial market for the system IP products that Arteris specializes in. With that, I will turn the call over to the operator and open it up for questions. Operator.

Operator (participant)

Ladies and gentlemen, we will now begin the Q&A session. Should you have any questions, please press star followed by the number one on your touch-tone phone. You will hear a prompt that your hand has been raised. Should you wish to decline from the polling process, please press star followed by the number two. If you are using a speakerphone, please make sure to lift your handset before pressing any keys. One moment while we prepare our Q&A roster. Your first question comes from the line of Josh Buchalter from TD Cowen. You may now ask your question.

Josh Buchalter (Senior Equity Research Analyst)

Hey, guys. Thank you for taking my question. Congrats on the steady results and an interesting backdrop. I guess I wanted to hit on the tariff and the trade environment right now. You're being upfront with sort of your cautious view. I guess I wanted to ask more directly, are you seeing any changes from your customers' behaviors yet, and in particular on their willingness and desire to invest in their IP and SoC roadmaps going forward, or is no change yet and you're just being more cautious on the back half?

Charlie Janac (CEO)

I mean, semiconductors, so IP and ADAS is not really subject to any tariffs directly, right? We are seeing some replanning of projects in China due to not only tariffs but also some of the increasing regulations coming from the U.S. On the other hand, what we're also seeing is that as the larger companies want to, they're looking at their efficiencies, and there seems to be an increased willingness by those companies to outsource system IP to commercial vendors such as Arteris. The Arteris type of companies actually tend to do fairly well in markets that are just under uncertainty because our customers tend to design their way out of recessions. While there may be some impact on royalties, which are a relatively small portion of our revenue at this point in time, the licensing activity remains robust.

Nick Hawkins (CFO)

Thank you for all the color there. And just to add to that, Josh, thanks for the question. This is Nick speaking. So yeah, just to reiterate in case anybody missed it, while we're pointing out that there is some economic uncertainty around because things are changing day to day, as you know, the business is still robust. We're not seeing a change in the pipeline one way or the other. We think we're well set for the year, which is why we haven't changed our full year midpoint guidance on any of the metrics. There is, as you rightly point out, a great deal of uncertainty still. We don't know what the knock-on impact and what the collateral impact might be, which is why, in common with many other companies, we've just widened our guidance range because we have less certainty right now.

Josh Buchalter (Senior Equity Research Analyst)

Okay. Makes sense.

Thank you for the color, both of you. For my follow-up, I wanted to ask about FlexGen. It sounds like you're increasingly confident in the success of that platform. Could you maybe elaborate on what's driving the expectations for revenue and ACV in the second half of the year and where sort of you're seeing initial traction from an application standpoint for FlexGen? Thank you.

Charlie Janac (CEO)

We have been working on FlexGen for three years. We have been delivering it to customers since the second half of last year. We have right now, at the moment, about 20 projects evaluating FlexGen. The feedback has been almost universally positive, and we have been able to get it to full production status back in February of this year with the release of FlexGen 1.2, which is the third release of it. We anticipate that it is going to generate substantial bookings and then revenue, which is ratable, in the second half. We are very pleased with the reception of FlexGen. The applications are fairly broad, right? They range from automotive, data center, enterprise, consumer. It is a pretty broad spectrum of companies that are using it for evaluations right now.

Nick Hawkins (CFO)

Just to have a little bit of color to that as well, Josh, as I'm sure you know from previous commentary, any increase from FlexGen, which is likely to be through its ISP as much as major new customers, is likely to be first or will be mathematically felt first in RPO and ACV because they step up contract by contract immediately. The revenue, because of its rational nature of everything, is running on a sort of two and a half to three and a half year contract design term because it's rational. The impact on revenue from any second-half deal signed or even second-quarter deal signed will be much more muted. We will see the full effect in 2026.

Josh Buchalter (Senior Equity Research Analyst)

Thank you.

Nick Hawkins (CFO)

It also may benefit free cash flow, of course, but to the extent that customers pay upfront.

Josh Buchalter (Senior Equity Research Analyst)

Thanks, both. Appreciate it.

Operator (participant)

Ladies and gentlemen, as a reminder, if you would like to ask a question, please press star followed by the number one on your touch-tone phone. If you are using a speakerphone, please make sure to lift your handset before pressing any keys. Your next question comes from the line of Kevin Garrigan from Rosenblatt Securities. Please ask your question.

Kevin Garrigan (Senior Research Analyst)

Yeah. Hey, Charlie, Nick, congrats on the solid results. Hey, you guys spoke about accelerated interest from companies that have insourced. Are you also seeing accelerating decision timelines by customers that are kind of going from having an initial conversation to signing the licensing agreement than you've previously seen?

Charlie Janac (CEO)

Not really. I mean, people are generally trying to accelerate their design cycles, but the licensing activity is really pretty much steady, I would say. Steady growth. There is, of course, a number of large companies that are looking to basically what we're seeing is that they're saying, "Okay, for whatever we're doing today, we're going to keep the existing system IP, but we're not going to invest in it for the next generation because it's becoming more complex and more difficult." The next generation tends to get outsourced, right? We are pretty pleased with that area of the business. We think ultimately, because of the complexity and new application, most of the system IP is going to ultimately get outsourced. We are going to go from maybe two-thirds of the market being internal to maybe one-third of the market being internal and two-thirds being commercial.

I think this economic uncertainty is actually accelerating that, even though we're not seeing much impact on faster sort of decision-making on individual projects.

Nick Hawkins (CFO)

Kevin, thanks for the question. This is Nick. Just to add to Charlie's excellent commentary, there are two key things that are driving that. One is a pure economic one, which is driven out by the complexity, the increasing complexity of SoC designs. It's becoming tougher and tougher to support internal teams, which are very expensive. Typically, we look at sort of a 10x payback from taking an Arteris license versus doing the design internally. Pure economics in times like we have at the moment, that is something that our customers we're noticing are supremely focused on, is improving their OpEx and improving the bottom line. The second is actually just a scarcity issue. We've touched on this before. There is a scarcity of qualified hardware engineers who are capable of designing networks on chip.

That is something that actually FlexGen plays into very well because it decreases the amount of sophistication required by a customer's design team to be able to design a knob. Those two factors are quite interesting and almost inexorable drivers of the move to the commercial market.

Kevin Garrigan (Senior Research Analyst)

Okay. Great. I appreciate the color on that. As a follow-up on your joining the Intel Foundry Alliance program, I know it's only been two weeks or so since the announcement, but have you guys seen any increase in interest? Do you think this, by joining the alliance, could help you land a customer or two that you might have otherwise thought you would not be able to capture?

Charlie Janac (CEO)

Yeah. I mean, a lot of this is related to Lip-Bu Tan taking over as CEO of Intel, right? What we're expecting is that Lip-Bu is going to be much more open toward commercial solutions, both EDA and on the IP side. Also, there is a commitment by Intel toward continued investment in the foundry technology and in customer relationships for the foundry. We want to make sure that we participate in the 18A sort of process deployment and that we're basically the main system IP partner for that alliance. Yes, we expect that this will result in some additional business over the next 12 months.

Kevin Garrigan (Senior Research Analyst)

Yeah, that makes sense. Okay. Great. Thanks, guys. And congrats again on the results.

Operator (participant)

Your next question is from the line of Gus Richard from Northland. Please ask your question.

Gus Richard (Senior Research Analyst)

Yeah. Thanks for taking the questions. First of all, was book to bill approximately one and a quarter? That was kind of what I was coming up with.

Nick Hawkins (CFO)

Yeah. Gus, thanks. We don't disclose bookings. I don't really want to comment on that.

Gus Richard (Senior Research Analyst)

Okay. Always got to ask.

Nick Hawkins (CFO)

You've always got to ask.

Gus Richard (Senior Research Analyst)

Second question, Charlie, you joined two chiplet alliances, IMEC and, I'm sorry, IMEC and Intel. Just wondering, a couple of questions around chiplets. How do you see that evolving? Has the industry coalesced around UCIe standard, or is another interconnect standard competing with that? What is the, how long is it going to take for things to settle out so people can mix and match chiplets is kind of where I'm trying to get at.

Charlie Janac (CEO)

Yeah. Chiplets are already in production, but these are homogeneous chiplet SoCs where essentially it is one company, one process. You have multi-die chips like Intel Meteor Lake essentially shipping in very significant volumes. What is just now starting to be established is a heterogeneous chiplet environment where the chiplets are coming from different suppliers. They are coming from different processes. They have to be packaged in innovative ways. This is going to take a number of years, even though people are already investing today, which is kind of why we are very interested in this particular aspect of the system IP world because basically we make chips communicate on a single die, but now we have to make chips communicate across multiple dies or multiple chiplets in a more efficient manner. That makes the system IP much more complex and much more valuable, right?

We're very excited about that. In order to make this work, the industry and the whole ecosystem has to standardize in terms of some standards. Clearly, you mentioned UCIe, Gus, and that seems to be the communication mechanism for cache-coherent communication between multiple dies, right? We're basically trying to make multiple dies that have multiple processors look like a single programming space to the software. It's quite complex. You need standards like UCIe. You need standards like IP Exact, SystemVerilog. We expect that over the next couple of years the industry will coalesce based on a few standards in order to lower the cost of each individual project because the only way that this is going to really work economically is if these things are repeatable based on certain standards. We're committed to support whatever standards are emerging.

The whole ecosystem has to coalesce around the standards in order to make these heterogeneous multi-die chiplets SoCs an economic reality.

Gus Richard (Senior Research Analyst)

Got it. And effectively, as this happens, you need a Switzerland NoC for the chiplets.

Charlie Janac (CEO)

Yeah. Exactly.

Gus Richard (Senior Research Analyst)

You talked about the market being two-thirds internal today, one-third outsourced. Roughly, what's your estimate on how big the market is today for system IP?

Charlie Janac (CEO)

Okay. So it's about somewhere between $1 billion and $1.2 billion. The NoC itself is about $700 million, something like this. There is probably $300 million in the SoC integration software. This is the software that's used to package, connect, and configure the various IP blocks. Then there's about probably another $300 million or $250 million in system IP blocks that do not do any processing. They're used to essentially make the chip function better and communicate better, more efficiently, and those kinds of things like last-level caches, memory controllers, and so on and so forth, reorder buffers, and things that just manipulate the traffic around. It's about $1.2 billion. $1.2 billion.

Gus Richard (Senior Research Analyst)

Got it. Got it. Super helpful. The last one for me, you talked about your large customers considering outsourcing. Obviously, the next generation will be more expensive, more complex, limited resources, etc. My question is, is there some underlying technological change that is making this happen now rather than two years from now or two years ago? What is prompting people to, "Okay, we got to do this"?

Charlie Janac (CEO)

I mean, when Arteris started, which was more than 20 years ago, you had one processor and two channels of memory and a couple of IOs, right? It was relatively easy to do this in-house. Then you started getting multi-core processors. You had to introduce cache coherency. The next revolution came with AI. You had these big AI sections, which have their own internal and external data communication requirements. They require very high bandwidth and so on and so forth. Now you have the rise of chiplets. With each of these, it is becoming more expensive and more difficult to do. There is not one single thing where the CFOs and the SVPs of engineering of these big companies wake up and say, "Oh my God, we got to make a change." With each of these developments, it becomes really, really complex.

I know, Gus, you went to the Intel Foundry Day as well. There was a Kevin O'Buckley slide which talked about a chiplet system that was 12 times the size of what's possible with reticles, right, which is about 600 sq mm. These were chiplet systems 12 times the size. When you start to deal with these kinds of monsters, it gets very, very expensive. That is what drives the adoption of commercial IP solutions, which can be amortized over many more projects than just one company can do.

Gus Richard (Senior Research Analyst)

Right. Last question for me, and I'll step back. You're in a certain position. It would take you so much R&D to get to what's needed for 12x the reticle size. Sort of what would it take a large company that does huge SoCs to develop something for the chiplet economy?

Charlie Janac (CEO)

I mean, these large companies, they have very large budgets, right? They can decide to invest in sort of the chiplet infrastructure. As Nick pointed out, the problem is getting the people, keeping them together for a number of years, and be able to amortize that investment over relatively few projects. I mean, most of these companies are doing two to three chiplet projects right now, right? They're not doing 10 or 15 like they're doing single-die projects, right? The cost, just the CFO of those companies looks at it and says, "Why are we doing that?" Companies like Arteris have a faster learning than any internal team. We can amortize that investment, which is starting to be quite large, against many, many more projects.

Faster learning and basically amortization of the R&D over many more projects is the advantage that we would have against any internal team. It is steady progress in this outsourcing to commercial solutions. It is not going to happen overnight, but the trend and the direction and the vector is clear.

Gus Richard (Senior Research Analyst)

Got it. Thanks a lot. Thanks for your patience.

Operator (participant)

Your next question is from the line of Blayne Curtis from Jefferies. Please go ahead.

Blayne Curtis (Senior Equity Analyst)

Hey, guys. Thanks for letting me take the last question here. Charlie, I just want to be curious what you would point to in terms of your annual forecast. You're looking for continued growth on the licensing side. I was wondering if you could talk to it by an end market perspective where you're most hopeful of for the remainder of the year.

Charlie Janac (CEO)

Clearly, the highest growth at the moment is in AI. There are AI projects for data center training. There are AI projects for data center inference. There are projects for edge inference. Actually, AI is making it into the endpoints like phones and cars. That is a large number of projects. I think probably about half of the design starts in the quarter were actually somehow AI-related, right? We continue to be bullish about automotive. The automotive industry has some cycles, but people are designing now automotive chips for six or seven years from now that are going to, so that is also pretty robust. We are very excited about our microcontroller entry. The microcontrollers are now becoming much more complex, so they need network on chip and software that helps the microcontrollers come together more efficiently.

There are the usual consumer and enterprise applications, which are also steady, right? What is really happening, Gus sort of asked about the market size, right? There are probably about 600-700 SoCs that are being designed a year. We have only 15% of that, right? There is plenty of growth. Each project uses more and more system IP, which becomes more and more valuable, right? We are very bullish about the opportunity that we have in front of us. Our main challenge is to make sure that we remain the main independent neutral player in this particular piece of the world.

Blayne Curtis (Senior Equity Analyst)

Gotcha. Thanks. Just for Nick, I wanted to ask you on the OpEx. You were, I think, last quarter talking about keeping SG&A flat. Obviously, it is at a higher level in the first half. Just kind of curious if flat is still the right way to think about it.

Nick Hawkins (CFO)

When we talked about full-year guidance last time around, Blayne, we characterized or I characterized 2025 as an up year by about 10%. Roughly half the rate of top-line growth. We have revenue growing this year guided at 18%. We are sort of maintaining our view on OpEx at about half of that, so 9-10% overall. I think that is still reasonable. Where we are getting all our operating leverage, and this is maybe what you are recalling, is in G&A, which has now been flat for three full years, including 2025, 2024, 2023 have been completely flat the way we are guiding it right now. Where we are putting the investment in is in R&D, sales, and field application engineering because they are all driving short, medium, and long-term top-line growth. We think there are huge opportunities left.

Blayne Curtis (Senior Equity Analyst)

Thanks, Nick.

Nick Hawkins (CFO)

Hey, Blayne, could I ask you just add on to Charlie's point? I was trying to cut in, but I realized I was on mute. Just to put some numbers around Charlie's commentary, AI-related deals now account for north of 55% of our total business. That is growing really nicely. That, of course, is a horizontal, not a vertical, as you know. The two highest growth rate verticals for us are, again, not surprisingly, automotive in terms of when I'm looking at ACV and royalties, automotive and enterprise. Enterprise is, of course, part of the AI story because that is obviously the backbone of the AI workload. That is very exciting. In fact, it is also starting, it is our fastest growing royalty segment as well, is enterprise.

Blayne Curtis (Senior Equity Analyst)

Thank you.

Operator (participant)

There are no further questions at this time. I'd like to turn the call over to Mr. Charlie Janac for closing comments. Sir, please go ahead.

Charlie Janac (CEO)

Thank you for your interest in Arteris. I hope that you're pleased with the results that we've delivered in Q1. We look forward to meeting with you at the upcoming investor conferences that we're participating in through the next couple of months. We look forward to updating you all on our business progress in the quarters to come. Thank you very much.

Operator (participant)

This concludes today's conference call. Thank you very much for your participation. You may now disconnect.