Before I hand over to Coby, just to note that we will be having a Q&A session at the end. If you have any questions, please type them into the Q&A box at the bottom of the screen. We're expecting a lot of questions to come through, so preference will be given to questions with a name attached to them rather than anonymously. I would now like to hand the webinar over to Coby. Please go ahead.
Hey, Danny, and hi, everyone. Good afternoon. I'm glad to have you all join our quarterly Meet the CEO. I think last quarter was really a great quarter. We've gone through a tough 2024, as everyone knows. We've been trying to get deals done with some of these big foundries, big IDMs for a long time. It's been a challenge, and I'd actually like to, before I focus on the last quarter, I'd like to go through a little bit of history and a little bit of what's going on and where the market is, actually, and many of you know it's a long time ago, over 15 years ago, the market already realized that Flash is basically going to hit its limits, and since these technologies take so long to develop, there were a lot of attempts to develop a replacement for Flash.
In those days, there was FeRAM. There was PCM with different variants of 3D XPoint and Optane, and there was MRAM, and there was CBRAM, and there was all kinds of technologies, and of course, ReRAM. Over the years, what we saw is one by one, these technologies kind of fell off the table. There's always someone who's still trying to make things work. But in general, the ferroelectric RAM, the PCM, and all the different variants of it, today, you might see one company trying to still make it work, but it's not really anything that the market is considering serious. The two last technologies standing were actually MRAM and ReRAM. Now, MRAM, five years ago, it was already in mass production. It was in a good position. I think the market basically assumed that's it. We found our winner. We found the technology that's going to work.
But as they started to manufacture, they realized it's very expensive, much more expensive than Flash, and definitely even more expensive than ReRAM. There are all kinds of other issues with it. It uses rare earth materials and things like that. But I think the key issue came up once these products hit the road and they were used in the market. The M, as everyone knows, stands for magnetic. And you'd be surprised how many magnetic fields we have around us. So some of these products ended up failing in the field. And that, I mean, they basically had the memory erased by some magnetic field somewhere. And that triggered the biggest consumer companies in the world to basically say, "Hey, our products are going everywhere. We really can't control where they are.
This is a big risk," and they passed a very strong message to the market that they're not going to use MRAM. Now, the way that this was reflected in the market was just a year and a half ago when TSMC released their NVM roadmap. It was still all MRAM with ReRAM just as a side note, but last June, it was basically updated, and now their NVM roadmap is all ReRAM with NVM as a side comment. And I think that reflects the whole transition in the market. We've gone a long way, but we finally reached a point, and it's really last year, that the market understands ReRAM is really it. It's not like MRAM is going to disappear. MRAM has its place, and people have already invested, and it's going to be used, but ReRAM is much cheaper.
It doesn't have all of these issues with magnetic fields and so on. It has the long list of advantages that we've talked about so many times in power and endurance and all these other things. And now it's ready for mass production. TSMC already has it in mass production. Weebit has it qualified, both at Leti, which is basically STMicro's wafers, and at SkyWater. We have agreements also with DB HiTek. We demonstrated it on GF wafers. So the market is basically waking up and saying, "Wow, this is the technology of choice. This is where we're going." And Weebit has a good technology. And last year, we also demonstrated it already working at high temperatures, and we demonstrated high endurance and so on. And everything was basically lining up.
We are, if you look at the ReRAM space during these years, the other ReRAM players, if it was the other smaller companies like CrossBar, Adesto, et cetera, they basically disappeared. If it's the big players, it's not a secret that even Intel and Samsung tried to develop ReRAM and ended up giving up. ReRAM is not an easy technology to develop, and you really, really need a very strong team that will develop it. You need the strong device guys and process guys and analog and digital and test and characterization and everything. And you need them all to work very closely together. You need to have a very focused management. That's where the Intel and Samsung guys had their issues.
You really need to have a very focused management that is driving this, that is working together, that is defining what to do with trade-offs and so on. And Weebit really is, I think, the only company in the world which has all of these things together with that big focus. This is what we do. This is really our big focus. I mentioned it several times. We have 13 PhDs. Actually, the number is going to grow soon, focused on developing this technology. And today, we are the only independent supplier of ReRAM. So last year, and even before that, in 2023, we were already engaged with so many of these big foundries, big IDMs. And they were evaluating and testing and even negotiating deals. But there was always that big concern about it's a new technology. It's a young company. They never showed their technology in mass production.
And so everyone stayed on the sidelines. Yeah, it was driving me crazy. You would see that everyone was asking me, "Where is the deal? Where is this big deal? You're saying that you're talking to all these big guys, and where is that deal?" Well, we were, and we are, continuing to work with these guys. But for them, it was a really big perceived risk. And so they were trying to do everything they could to continue to work with Flash and wait for someone else to make this step. Now, we ended up with last quarter. I think it was a great quarter for us. It started with the Semiconductor Australia Conference, which I really enjoyed going to. It continued with the raise that we did. I have to say, an unexpected raise kind of.
I wasn't really planning to do a raise, but when a broker comes and says, "I have all this money underwritten and at a premium over market," et cetera, I think it was a good raise for us. It really gave us now several more years of quiet so that we can really focus on doing the work. And at the end of the quarter, as everyone knows, onsemi ended up being the first one to cross the line. They announced their Treo Platform, which is a very nice advanced technology at 65 nm BCD. I mean, you go back, I don't remember when exactly, but several months ago, they had that big announcement. And they need advanced memory. That's when you announce a new platform and everything already moving forward, you don't go back to a technology like Flash. And so we had talks with them.
Of course, just like everyone else, they had their concerns, but I think they saw that Weebit has such a strong team, is capable of really resolving any issues that will come up, et cetera, and I'm very glad to say we signed this agreement. This past month since we signed the agreement has been phenomenal. They are so focused. This is such an impressive company. They are taking this project so seriously with so many people involved, and we're running forward in parallel with the device and the process and the design side and the test characterization, everything. We have so many meetings and so many things going on in parallel. It's been great, and in addition, of course, the impact on the market has been great. The way that the market perceives Weebit now has really changed.
Last week, I got a phone call from a VP in one of the biggest foundries. I always call him up. Yes, we have good discussions. Yes, his team is really interested in our ReRAM. Yes, it's been more than a year that we're talking and analyzing and checking, and they're hesitating, and suddenly, he calls me up and says, "Coby, you need to send us a plan. How do we ramp up on ReRAM?" That's like, "Wow, suddenly, they're pushing instead of me," or they're pulling, if you want to look at it, so it's that kind of thing. I have so much travel now planned to meet so many of these foundries and IDMs, et cetera, so the way that people are looking at Weebit is really changing. People are obviously now someone already took that first step.
So it's not like everyone is rushing and everyone's now standing in line to sign the agreements, but they're definitely moving forward and things are moving easier. We were at the CES conference in early January. CES is a huge conference. I don't know if you can imagine a conference where 150,000 people get together in one place. It's like a whole city just coming together. And we had amazing meetings there with people in the industry. I got to meet, of course, a lot of the onsemi guys there, the CEO of onsemi, et cetera. But other customers, our potential customers, to be exact. We met many bankers, the leading bankers in the world were there. There was a lot of excitement around it. I had one analyst, actually, last week, tell me, "For so long, there were so many players in this emerging NVM space.
I've been following it for so many years. And suddenly, it's Weebit." So it's exciting times. I'm really excited about onsemi. They're an amazing partner. We're continuing, of course, to work with DB HiTek, and they are a great partner. And we really are enjoying the work with them and moving forward. And I think 2025 started with a blast, and there's so much going on. It's known that my entire equity compensation for this year is tied to closing deals. And I definitely plan not only to achieve my target, but to even go over it. And I think there is a potential for that. We're very focused on this. The team is all excited and ramped up. So yeah, I think 2025 is going to be an amazing year. Yeah. So I don't know.
I think at this point, I know many people have questions, and I'll open the floor to questions. I'll be glad to answer.
Sure. Thank you, Coby. We'll now move to the Q&A session, as Coby said. Look, once again, if you have any questions, type them into the Q&A box at the end of the screen. We do have a healthy number of questions coming through, and I will try and ask them all. A reminder, we will be giving preference to non-anonymous type questions. Okay, Coby, the first question is around GlobalFoundries. So can you maybe just give a brief update on Weebit's dealings with GlobalFoundries?
GF is an interesting case. On the one hand, as everyone knows, we manufactured our ReRAM on their 22 nm wafers. We're continuing, by the way, to manufacture there and to test and improve the technology.
In parallel, and I think everyone knows already in 2023, they announced that in the beginning, they acquired what was left of Adesto's CBRAM technology. And there has been a pressure internally in GF to try to develop ReRAM on their own. So it's kind of going in both directions. I think at this point, Weebit has a lot of potential with a lot of foundries and IDMs. GF is one of them. The others are very focused on working with Weebit. With GF, it's not really clear where this is heading. So we're continuing to manufacture there. But yeah, I admit my disappointment that they are still trying to develop their own version, and we'll see what happens there. I don't know where it's going to go.
Okay. The next question is about the recent capital raise.
Coby, you've now secured enough liquidity for a few years of runway following that capital raise. What continues to drive you and the leadership team, and what keeps you motivated?
Oh, wow. I've never been so full of energy. I think this is the point in time when you look forward, and it's just amazing. I mean, I couldn't have dreamt of such a situation. There's this huge vacuum out there. The market basically decided it wants ReRAM. ReRAM is the future. Yes, a lot of people are still hesitating. This is a very conservative industry, believe it or not, because of the huge amounts of investment. So companies invest billions in building these fabs and in everything. And so much is at stake.
Even the Treo Platform, when you think about it, when they made the announcement, they kind of said in their announcement that it's a trillion-dollar type of project for onsemi. And so people are not rushing through. But at the end of the day, there's this huge vacuum. Everyone knows they want ReRAM. They need ReRAM. Everyone knows they're going to be using ReRAM. And Weebit right now is the only independent supplier of ReRAM. And because it takes so long to develop ReRAM, you can look around and see that I basically cannot see any other independent company that is going to have ReRAM anytime soon. Someone might have it. It might be in stealth mode or whatever. It's very hard to keep these things as a secret. So right now, Weebit really needs to rush in and go and take as much as it can of this vacuum.
The nature does not support vacuum. Wherever there's vacuum, it has to be filled. And we are now so totally focused on we have to fill this vacuum. There is so much. There are so many of these big foundries, big IDMs, these product companies that need a solution. And we have it, and it's just we need to overcome this inherent resistance of people being afraid of working with a younger company, with new technology, et cetera. But as we break this down, as we now have DBH, we have onsemi, we'll sign one or two others, I believe that that level of resistance will fall. And I mentioned in a couple of meetings the book Inside the Tornado, or I'm sorry, Crossing the Chasm. That's the first book in the series. It's a great book that I recommend people to read.
It basically talks about how young companies that develop new technologies evolve. And I'm going to skip a lot of it, but there's the phase where you have this crazy idea for a new technology, and you're developing this, and then you finally have it ready, and you want to go out and sell it, and you find out that there's this huge chasm, and all of the customers are on the other side. And you go and you eventually manage to cross this chasm, and you're coming out the other side. And in the beginning, you enter what's called the bowling alley. You basically manage to find one company that already really needs this technology and sees that the potential benefits are so much higher than the perceived risk. And they move forward with you.
You have this DB HiTek. You have onsemi that realized just the huge potential of this technology and that it is something that they should go with. The first pin drops two pins behind it, and then those drop the three pins behind them and so on. The book describes it in a nice way. Once you have a few pins falling down and you're starting to make that progress, there's what people call today FOMO, and suddenly, everyone is there like, "Oh my God, my competitors have it. Everyone's moving forward with it, and I'm going to be left behind." Suddenly, you go into the phase which is called Inside the Tornado. Suddenly, everyone wants the technology. Now, I've been fortunate enough to be inside the tornado twice in my life. It is the most amazing feeling.
It's unbelievable what happens when suddenly you're swamped with people calling you up and wanting the technology, and everything just starts growing so fast. Now, I can see that tornado out there. I can see the tornado out there. Could be 2026 or so. This year is still going to be a very hard year of going through that bowling alley, of getting a few more pins knocked down, and we're making great progress, and the onsemi announcement already changed the whole atmosphere around Weebit, and we're going to get into that tornado, and that's, I mean, just waiting for that, not waiting, I mean, pushing for that to happen and running forward towards that tornado is really what's driving me, and it is so exciting, and I think the whole team feels it. The team is working around the clock now. It's amazing.
The amount of travel that I have planned in the next couple of months, I'm hardly going to be at home. My wife is going crazy. But I mean, that's part of starting to move in that direction. It's amazing.
Fantastic. Okay. The next question is around the size of the memory arrays. It's a rather long question, so I'll just truncate it. So basically, what the question is, and I'll put context around it. So many devices on the planet need much more than 1 MB or 8 MB of NVM. Has Weebit produced a memory array bigger than the 1 MB?
So we need to make a clear distinction between the embedded memory space and the discrete memory space. You're absolutely right. On the discrete side, we're using so much memory and AI and the huge data centers and whatever they need, giga, tera, and beyond.
But in the embedded space, actually, even one megabit is considered a lot. A lot of the devices need 256 kB or 512 kB. Weebit already has an array of 8 Mb , which is very impressive, and people are very impressed by it. We showed it last year at the Embedded World Conference, and there was a lot of excitement around it. Most of the requests that we are getting are below the megabit level or at the megabit level, let's say. And we're feeling very good with it. We can go and develop even larger arrays. And some people have been asking us about larger arrays. So I won't be surprised if we go ahead and with one of these people end up developing something bigger. But in the embedded space, megabit is already considered a large array and something that most people are very happy with.
Yeah. Okay.
Very good. The next question is around remuneration, specifically yours, Coby. So can you confirm that your salary bonus this year is conditional on a number of signings by fiscal year or calendar year?
So my salary consists of three parts. There's the base salary. There's what we call STI, which is target-based bonus. And there's the equity part, target-based quote bonus as it's defined. Now, we work on a calendar year. It's just so much easier. That's how the industry works. So we just need to work with the industry. So my targets are calendar year. And we just actually, the board is, as we speak, the board is discussing my targets for 2025 and closing those. And so that's in terms of the target-based.
By the way, the vast majority of my compensation is target-based and equity-based, which is intended to motivate me to meet these targets and to want the share price to go up, et cetera, et cetera. The equity, as the shareholders know, is tied this time 100% of it, which is, by the way, I have to say, it's very, very unusual to have such a situation where the equity is tied to milestones, not to mention to 100% of it. The board, knowing that everyone is so anxious to see deals signed, we had a discussion, and I volunteered to actually have my equity compensation for the calendar year 2025 be tied to closing deals. I definitely intend to meet that target. As I said, I actually want to overachieve that target. I think we can.
Okay. The next question is we might turn to your end customers. Maybe a couple of questions on your end customers. So one investor's asking, "Three end product customers by the end of 2025 calendar year. Doesn't sound like too many."
You need to remember that these end customers need a fab to manufacture it. And that's the big challenge. Actually, for me, the toughest and most difficult challenge of them all is these product companies. They will not really move forward without a fab where they can manufacture. Unfortunately, as I think everyone knows, SkyWater has made a strategic decision to focus on their R&D services. If you look at their quarterly report, you'll see that by now, they already got up to a point where 90% of their revenues are from R&D services and not from foundry. It's their decision.
Very disappointing on my side. So getting a customer to manufacture in SkyWater is a huge challenge. I'm not saying that we won't have it, and we are working on trying to get customers also for SkyWater, but it's very hard to actually get someone over there right now with the current focus of SkyWater. DB HiTek, we are working on qualifying this year. We have wafers from DB HiTek that we are testing now. We'll need to get to the point where we can actually start a qualification and then do the qualification. And as everyone knows, we're talking about many months of putting the chips in ovens, et cetera, et cetera. So again, most customers will not want to engage with us before we are actually qualified or very close to qualification.
So the target of getting product companies is the most difficult one that I have right now on the list, both in my target-based bonus and with my equities. But I want to make it clear when I have a target, I achieve it. I focus on it, and I will make it happen. And we will get product companies on board and working. Of course, onsemi is great because it's an IDM. So I really believe that we'll have a product. We're trying to work with them now on defining a product that will use our ReRAM. So I hope to have that as a first product, but we still have two others that we really need to work on. And again, my goal is to overachieve.
Yep. So I'll ask the end customer question another way because follow-ups come through.
So if the future is ReRAM, Coby, end customers haven't signed up yet to get a first mover advantage, but it sounds like some of them have gone through TSMC. So why can't you get the signatures?
Well, so first of all, we need the fabs. I think that's something that I already answered. And even TSMC, look at the ramp-up of ReRAM. In the beginning, everyone is hesitant. Everyone is moving a little slower. But we've had, and we are having, and we're going to have a lot of meetings with product companies. I mean, just yesterday, I think the company had one or two meetings with product companies. Next week, I'm going to be meeting some CEOs. And next week, the next couple of weeks, I'm going to have several meetings with potential product companies.
It's happening all the time now and more and more so, actually. Again, the onsemi announcement really triggered, and by the way, onsemi, while it's an IDM, it does give some foundry services. And we are also engaged with people that are interested in, I mean, maybe manufacturing at onsemi.
Yep. Speaking of onsemi, we'll ask a couple of questions around onsemi. First one is, which particular product line has been decided with onsemi?
It's under discussion, and it's not public information, so I can't really go into that. I will make a general comment, and this is just really a general comment and not an indication of the product, but onsemi is well known for two main application areas. They are well known for power management, and they are well known in the automotive space.
I think the key reasons for them to engage with Weebit were that they saw that we are very strong in both of these domains and give them strong advantages. So that was one of the if people ask, "Why did onsemi make that step forward?" It's because on the power management, and we've been talking a lot about power management recently, that's kind of the really low-hanging fruit where we have so many advantages with ReRAM that it's really compelling for people to move forward. Then the fact that we demonstrated that we can work at 150 degrees, which is really the main requirement for automotive. These two things together, I believe, if you ask me, I believe that those were the main considerations of onsemi when they decided to move forward.
That segues beautifully into the next question, Coby.
So onsemi holds the number one market position in auto sensors. You mentioned that was one of their two major areas being autos. Apart from power management and the 150 degrees, what are the competitive advantages of ReRAM that make it so applicable to autos specifically?
So to autos, it's really the power more than anything else. I'm sorry, the temperature more than anything else. And by the way, we have people asking us about even higher than 150 degrees. We have people asking us about 175 and so on, which is really extreme already. I want to make it clear. There's really not a lot going on above 150. But we basically showed the simulations and the analysis that we can achieve even 175. And by the way, once you go above 150, you stop talking about 10 years or this.
You start talking about usage profile and how long does it have to stay at that high temperature and things like that. A small anecdote, we had a mining company that needs chips in their drills, and those drills get really hot. And they said, "We need something that can operate at very high temperatures, but it doesn't have to be for years. Even a day or hours would be good enough. We'll replace it." So you get to that point. But the temperature is a key criteria in the automotive space. onsemi, with their sensors and so on, they need that. And they also, as I mentioned, the power management side of things and the fact that with ReRAM, you can have a one-chip solution instead of a two-chip solution, which is much more efficient and much lower cost, et cetera. I think those were the main criteria.
There is a follow-up question also on onsemi. Okay. So post the onsemi announcement, why do you believe the market reacted so poorly? Do you think the market expected large revenue streams as part of the announcement?
Well, the market in general, in the Australian market specifically, is something that I never managed to predict. And it doesn't stop to even more than surprise, astonish me. I don't know what the right word is, but I just remain in awe sometimes. I really don't know the considerations and what drives the market and what makes people decide to buy or sell. In these kinds of deals, you can't and you don't want to. And I don't think the shareholders want us to give a lot of the commercial terms. At the end of the day, I think it's a great deal for Weebit.
I think we have good terms. People will see during the year, as the year progresses, the cash flow, the revenue, et cetera, implications of this deal, and so people will realize that it might be that because we didn't announce commercial terms, people thought that this might not be a significant deal or anything like that. I don't know. You have these situations. I mean, onsemi, let's face it, these big guys don't even like to have the press release. The fact that we actually had a press release is already a big achievement and a big success. Going into commercial terms is something that is just not done in these kinds of announcements. Throughout my, let's say, on the sales side, over 30 years already, I've never done a press release where there were commercial terms in it. Doesn't make any sense.
I think the people who would be hurt most of it would be the shareholders because once you announce commercial terms, you basically put a glass ceiling on your head, and no other customer will ever agree to pay anything above that. They'll all use that as the starting point for negotiations. So it doesn't make any sense. So I think I don't know why the market reacted this way. Maybe because the share price already went up quite a bit before that. Maybe the market anticipated it. I don't know. But Weebit is going forward. There will be the cash coming in. There will be the revenue coming in. People will see the impact of this deal. I believe that I'm a long-term person. I don't believe in the one-day type of things. And people will see the impact.
They'll see the additional deals that are going to happen. Hopefully, the market will reflect it. Absolutely. By the way, I want to comment on something. I saw this ridiculous comment somewhere, and I just have to comment on this, and by the way, it happened to me with DB HiTek as well, so I had someone ask me after the DB HiTek deal, "So how much did Weebit pay DB HiTek to close the deal?", and now I saw a comment. Again, I can't believe the ridiculous comment of, "So Weebit probably paid onsemi to do the deal.", so I just want to make it extremely clear. onsemi, DB HiTek, they are the customers. We're the vendors selling to them. It doesn't make any sense anywhere in any form for us to pay them anything.
I can't even believe people think that we actually paid anyone for something like that. So I just want to make it extremely clear. These are commercial deals where Weebit is the vendor. They are the customer. We're selling them or licensing our technology to them. And that's it. Weebit does not play any kinds of absurd things like that. And maybe people thought that, "Oh, Weebit is using SkyWater to do some R&D things." So I think I mentioned in the past that we have two separate agreements with SkyWater. One is a foundry, and one is an R&D center, which is a unique situation. Other foundries are just foundries. SkyWater, for some reason, decided to focus on giving R&D services. Weebit is very happy as an R&D customer of SkyWater. Weebit is using SkyWater, I can even say intensively, as an R&D center. They complement Leti very well.
You could say that the R side is done at Leti, the D side is done at SkyWater. But that's a unique type of agreement. So with SkyWater, we have an agreement as a foundry and then a separate agreement as an R&D customer. DBH, onsemi, they don't do R&D services, and we have just simple, straightforward agreements with them as foundries, and I wanted to be very clear. We didn't do any funny games there.
Yep. No, that's crystal clear, I think, Coby. Just speaking about DB HiTek, we do have three or four questions on DB HiTek, but I'm going to put them into one or maybe two because they're all asking pretty much the same thing. So the target for qualification for DBH was June 2025. Is that target now sometime in 2025?
Yeah, I think right now I'm saying sometime in 2025 just because these things are very tough, and we have the first wafers that came out. Actually, we have very good results on these first wafers. The tests are looking good, and we're working very closely with DB HiTek to push this forward, so I believe the qualification will be definitely before the end of the calendar year, I'm sorry. But I think that putting a specific target right now will not serve us, will not serve anything, so we are focused on finishing it as soon as possible, and it's hard to say right now when exactly it'll happen, so we're saying 2025, but the goal is definitely not the end of 2025 and to try to get it done much before that.
Maybe a follow-on from that and looking a bit more long-term. So how long after qualification do you think it will take to close the first customer agreement?
Oh, that's the billion-dollar question, right? We are engaged today with customers that are interested in manufacturing at DB HiTek. We even know that DB HiTek salespeople have already talked to customers about Weebit. So this is happening. The product companies, obviously, they want to see more progress. And I mean, they ideally would like to see qualification before they engaged. By the way, that's one of the reasons why I really want to finish the qualification ahead of the end of the year is so that we have enough time after that to close agreements with the product companies.
Going back to the previous question on product companies, so my goal is to finish the qualification definitely before the end of this year so that we have enough time to close product company agreements as well at DB HiTek. We do want to close agreements with product companies even before qualification is done. These are very tough things to predict. We're doing our best. That's what I can say.
Yep. Thank you. There's a few questions also around the discrete selector development. So again, I'll just collate them into the one question. So can you maybe just give an update on the progress with regards to the development of a selector for discrete ReRAM memory?
So Weebit is working on this. There's always some work in the background. I have intentionally put it as low priority right now. We have so much going on on the embedded side.
I believe that in the embedded space, we have this huge potential. I talked about this big vacuum that's formed in the market. We have so many of these big foundries and IDMs and product companies, and everyone is wanting ReRAM. It's kind of the thing that everyone knows they need it, they want it, and they're just hesitating. We need to push them over the line. I really want the team to be totally focused on, "Let's fill this vacuum. Let's rush in there and get it, get to that tornado, and really grow." I think that's the number one priority for Weebit. We are working. By the way, there are several projects that are always being dealt with in the background. Things like the discrete, that's one example. We have other projects, neuromorphic, and people know that that's always in there in the background.
We are working with research institutes on these things. So Weebit is now big enough that it can afford to have a small side activity on some of these strategic things for the future. But at the end of the day, I have limited resources, and I'm sure that the shareholders want to see the revenue ramp from the embedded and getting these big deals done. So that's really the big focus right now at Weebit.
Okay. You touched on SkyWater. There is one or two questions on SkyWater. Again, I'll merge them. So can you maybe just talk to the progress towards signing customers at SkyWater?
So as I mentioned, there's a strategic decision at SkyWater to basically move to be an R&D center. And again, 90% of their revenues now are already there. I remember last year I was talking about 75%.
I think you can see the big focus they have and how they're shifting the revenue from the foundry service. Now, almost all of it is coming from R&D. So they are lacking some basic elements in their foundry that we've brought. And I mentioned it in the past. We've brought some big-name product companies. I think I said in the past names that people in the street in Australia would even recognize. We've brought big companies to SkyWater who wanted to manufacture there with our ReRAM. But they were lacking some basic elements. It just ended up not working out. The whole thing fell apart, unfortunately, to my very, very big disappointment. But every company gets to define their own strategy and their own priorities. I haven't given up. I haven't given up. And we are looking for company.
We are talking to companies that potentially can manufacture at SkyWater despite all the limitations. And right now, I mean, right now, the place where we're qualified is SkyWater. I need to bring in three product companies, right? I need to try to make it work also at SkyWater. So we haven't given up on it, but it is a huge challenge because they're lacking some very basic elements in their library that companies look at it and say, "We can't manufacture there." There's basic IP that's missing.
Okay. Before I go to the next question, Coby, just a note to all the participants on the call. We've got about 15 minutes to go. We've still got about 10 questions to get through, so we probably can't accept any more questions unless they're very important. Okay. The next question, Coby, is around MRAM.
It's known that MRAM has some better performance characteristics than ReRAM, but TSMC has now adopted ReRAM. What's your ratio split between MRAM and ReRAM?
MRAM has two advantages today. One is it's just more mature. It's already in mass production. It's already out there. It's running. And obviously, that's an advantage. The second advantage is the endurance. MRAM manages to achieve higher endurance, and some applications require that. MRAM has its place. It's not going to disappear. By the way, in this industry, you always have all kinds of technologies that are used for niche needs that have some unique advantages. A lot of people are still using rotating disks in the data centers and even magnetic tapes as NVM memory. I don't know how to predict right now what part MRAM will have.
I personally, I'm tainted, but I personally believe MRAM will, over the years, the MRAM market share will shrink to specific applications and specific things. The numbers that analysts are bringing out are constantly changing also, and I haven't really seen a good analysis of it yet. But it is very clear now that the big focus, once TSMC moved all of its focus or the vast majority of its focus to ReRAM, Weebit is here pushing it. You can see that GF is trying to make, for some reason, they decided to try to make their own version, and we'll see where that goes, so I think ReRAM will clearly take the majority of the market. How big that majority will be, I personally think it'll be a large majority, but I think the jury is still out.
Okay. There's a follow-up question on discrete.
So Coby, is there still an intention? Is there still an intention of developing a smaller discrete ReRAM module that doesn't require a selector?
So let me correct that question and use the right terminology. In ReRAM, you have to have a selector. What we're trying to do is replace the transistor with a smaller selector. So that is the goal of our discrete project. The ReRAM memory bit itself is very small. And because every bit needs a selector, and today we use a transistor as a selector, that's what makes it bigger. And that's what makes it not so ideal for discrete memories. So the ReRAM itself is good. We need to find a better selector, which will enable us to put the gigabits and so on on a chip. And that's what we're focused on.
So yes, we are looking for ways to have a smaller selector. This is a huge challenge. It is a very big challenge. Much bigger companies than Weebit have tried to find smaller selectors, more efficient selectors, and failed. There were big projects of very big companies to do it. Having said that, Weebit up to now managed to show that it can succeed even where the very big guys didn't. We are going to be focused on it. It is going to take us. This is a long project. This is a project of years. The day that we do it, it is going to be, "Oh my God," because then it really will open up an even bigger market. We are working on it. Eventually, we will find a good selector.
And by the way, of course, as you can imagine, we've been looking at different types of selectors, and we've been playing with this and going from one type to another and improving different things. So there's R&D work going on, a lot of exploration. And it's going to take us some more time. It's not a one- or two-year project. It's a long-term project. There is that work. Again, it's low priority in the background. I want the team to focus on this embedded opportunity. Now is a once-in-a-lifetime thing. And this huge vacuum that formed is amazing. We need to capitalize on that.
Okay. The next question is around ReRAM and the use in products. So a while ago, Coby, you talked about Apple or you mentioned Apple using ReRAM in one of their new iPhones.
Are you aware of any other products in the market that are using ReRAM?
We are aware. Some of them made it public. I think I mentioned Nuvoton announced, I think it was eight, nine months ago, that they have a new microcontroller that has ReRAM in it. They did a lot of marketing around the advantages of ReRAM in their MCU. I think that the interesting point that I would like to point out here if I'm talking about it is Nuvoton is actually the owner of the technology that UMC is using. UMC qualified a ReRAM technology and theoretically has it available. I'm not aware of any customer using the UMC ReRAM. And by the way, Nuvoton themselves use TSMC for this MCU.
Nuvoton announced that Infineon, a few years ago, already announced that all of the future automotive chips that they do will have ReRAM in them. So there have been some announcements around this. It's ramping up. The usage of ReRAM is very initial. The ramp-up of these things takes time. But it's definitely happening.
Okay. A devil's advocate question from a pretty tough investor here. So let's say worst-case scenario, Coby, you have no signings this year. At what point would you start looking at cost-cutting?
So let's start off with the fact that Weebit is extremely conscious of cash and of budget. Last year, because things took longer, I was really hoping that we'd managed to sign a big agreement earlier. As we saw that things are delayed, we were already cost-cutting, and we were saving budget.
I think the board was quite impressed by how much below the planned budget we actually were in 2024 because I don't hire people before I know that I have the customer. We need a team for every fab, and there's a very delicate balancing act between being ready, so when you sign an agreement with the fab, you have the whole team up and running and ready and everything, and not hiring people and then finding out that the agreement is pushed out three, four, six months, whatever, and you have a team sitting there idle. So we are extremely conscious about budget. Every PO, I can tell you just yesterday, there was a request for a PO for $2,000 that we looked at and we said, "Yeah, it's nice. It makes sense. But no, I'm not going to sign it.
We're not going to do it." Because we question every expense. The money that we have is money that we got from shareholders. It's not my money. It's the shareholders' money. They gave it to me in order to achieve a certain goal, in order to give them the maximum return. And we are very, very cautious. Even when we do travel, I keep asking, "How can we cut down the costs of the travel? How can we save money, either find cheaper flights or lower-cost hotels and so on?" It's all the time in our minds. It's not something that I need to wait for not having a deal done in order to think about it.
Right. We've got a few more questions to get through. The timing's pretty good. We'll maybe talk a little bit of some financial markets questions.
Now, it's pretty clear the Australian market doesn't appear to understand the microchip industry properly. Would you be contemplating or considering moving to the Nasdaq where it's very much likely you would get a much higher valuation?
Well, as I often joke, if I would get $1 for every time I'm asked that question, I would never have to raise money. It's the type of question that I'm asked all the time. And I'll tell you honestly, I mean, yes, the natural place for Weebit is Nasdaq. That's kind of everyone tells me, "What are you doing in Australia? Nobody knows semiconductor there." We are in Australia today. The Australian market, let's face it, people know that I have my issues with some of the regulation and so on. And I've had meetings with the ASX. We've been talking to the ASX about it and so on.
But at the end of the day, Weebit raised, I believe, close to AUD 150 million on the ASX. It's not like the ASX is bad to us. I love my shareholders. I think we have an amazing, amazing group of shareholders. I love the interaction with the shareholders and the fact that I can talk to them. And so many of them have been so loyal and stuck around with Weebit for so many years. So to be honest, I don't have a strong motivation to just say, "Oh, this is crazy. I can't stand this anymore. I have to go to Nasdaq." If and when and whatever this issue will come up, the board will be discussing it. It's not something that at this point right now, it's burning or a burning issue or something that I have to deal with now.
Do you think you're a takeover target?
In this industry, everyone's a takeover target. If you read the press, Intel is a takeover target right now. So I'm glad to be in the camp of Intel. No, but seriously, that's how this industry works. The big guys are constantly monitoring the small guys. They're constantly looking at how they develop. And as the smaller guys grow, the big guys, they constantly have these discussions of if and who we want to take over or even more like who we want to take over. And there's always acquisitions. I said it many times. I am totally focused on making Weebit a big company, a successful company. That is my sole goal. From my experience, and you can tell I've been around for a while, and my experience has been with startups. Most of my career has been with startups.
The moment that a CEO starts thinking about acquisition, being acquired, and so on, that's when the company starts going down. So this is not something that I am wasting any time on. I want Weebit to be big and strong so that the day that someone gives us an offer, I can say no if it's not a good one.
Absolutely. Final financial markets questions. And if we've got time, just a couple of follow-ups to some answers you gave earlier. So we'll be very quick. So the final financial markets question is, you met some analysts earlier this year with a view to potentially covering Weebit. Can you maybe give some update in terms of those brokers and where they're at?
Well, first of all, I want to thank MST, and I want to thank UCP for initiating coverage beyond, of course, Pitt Street that has been doing it for a long time. But MST really took the ball and ran forward with this and has been pushing it with their customers. And UCP, I mean, they did the raise based on the analysis that they did. Once the analysis was out, they had calls with institutions. That's how they got the money for the raise, right? So I've made it clear that I'm tired of having brokers promise to work with us, to give coverage, etc. They end up doing the raise, and then they disappear somewhere. And so my focus will be in the future, if and when we do raises, to focus on people who give us coverage, who are serious about Weebit.
I'm very glad, by the way. I don't know if people noticed, but Macquarie, in two different reports recently, mentioned Weebit. They weren't reports focused on Weebit, but they mentioned Weebit, so I think we're starting to see analysts, even in Australia now, more focused and looking to cover Weebit. I'm obviously going to be very glad to work with any analyst that wants to cover us, and hopefully, we'll get more and more coverage now.
Two clarifications, Coby, before we finish. Okay. The first one is around the onsemi qualification. So will they make their first set of chipsets specifically designed for the product in qualification?
I think all of that is work that we're doing with them now.
Obviously, I think the smartest thing, and this is the direction that we're trying to go in, is that the module that will qualify, the first module that will qualify, will actually be the module that will be used in the first product. It just makes sense both for Weebit and for onsemi. So that's kind of the direction that we're going in. Again, we're defining also what that target product would be, etc. So this is all stuff that's being defined now.
Okay. And the final clarification going back to the discrete selector, is it a smaller discrete module that uses transistors?
No, that's almost a contradiction in the statement. The challenge that we have, without going too much into technology, the challenge is that every bit needs a selector. The selector needs to supply a certain current and voltage and so on.
And the levels of current and voltage that the ReRAM needs are such that we are using transistors which are at a certain size, and those transistors are relatively bigger for the discrete market. Now, it's true that it is possible to do some discrete chips also with the current transistors. And I mentioned that in the past, and it's not something that we've dropped. There is a potential that we might have some discrete solution that is based on the transistors. We can be competitive even with the current transistor in some specific applications and domains. But for the vast majority of the discrete market, there needs to be a smaller selector, and it will not be a transistor. So yeah, we're working on, I guess, all of the above. But all of it is right now in lower priority.
And I would focus right now really on the embedded side. In the near future, in 2025, 2026, the big progress of Weebit, the big growth in cash flow, in revenues, and in deals is going to come from the embedded side.
Yeah. That's a great way to finish, I think. So that concludes the Q&A session. I want to thank all the participants for the healthy number of questions and the quality of them. Very, very well done. I will now hand back to Coby for any closing remarks.
Well, thanks, Danny, for holding this session. Thanks to all of the shareholders that attended. Just like I said in the beginning, I think we're looking at a really exciting year ahead of us, 2025 and then 2026. We have a unique opportunity. We're going to be filling this vacuum. We're going to be signing some additional deals.
The market is going to see the ramping of revenues and cash and so on. And I really hope to see the markets also react to it. So thanks, everyone. And let's have a great 2025.
Thank you.