Hesai Group (NASDAQ:HSAI) Q4 2022 Earnings Conference Call March 15, 2023 9:00 PM ET

Company Participants

Rachel Yang - Vice President of Operations
David Li - CEO
Louis Hsieh - Global CFO

Conference Call Participants

Olivia Xu - Goldman Sachs
Tim Hsiao - Morgan Stanley
Bin Wang - Credit Suisse
Paul Gong - UBS

Operator

Hello, ladies and gentlemen. Thank you for standing by for the Fourth Quarter and Full Year 2022 Earnings conference call for Hesai Group. At this time, all participants are in listen-only mode. Please note that today's conference call is being recorded.

I will now turn the call over to the first speaker today, Rachel Yang, Vice President of Operations for the Company. Please go ahead.

Rachel Yang

Thank you, operator. Hello, everyone, and thank you for joining Hesai Group's Fourth Quarter and Full Year 2022 Earnings conference call. Our earnings press release was distributed earlier today with newswire services and is posted in the Investor Relations section of our Web site at investors.hesaitech.com along with the webcast access to today's call. On today's call, we have our CEO David Li; and our Global CFO, Louis Hsieh. David and Louis will each provide their prepared remarks and we'll conclude the call with a Q&A section.

I like to remind everyone that our earnings calls and investor materials contain forward-looking statements, which are subject to future events and uncertainties. Our actual results may differ materially from those forward-looking statements. All forward-looking statements should be considered in conjunction with the cautionary statement in our earnings release and the risk factors included in our filings with the SEC.

Finally, this could also include certain non-GAAP financial measures. You should carefully consider the comparable GAAP measures. Reconciliation of non-GAAP and GAAP measures is included in your earnings release.

I would now like to turn the call over to our CEO, David Li. Please go ahead.

David Li

Thank you, Rachel, and welcome, everyone, to Hesai's first earning conference call as a public Company, I'd like to first express my gratitude to our investors, partners, dedicated employees and everyone who contributed to our successful IPO on February 9th.

During our IPO, Hesai sold 10 million ADRs at USD19 per share, and the underwriters partially exercised their over-allotment option, raising approximately USD192.4 million. Listing on the Nasdaq marks a milestone in Hesai's growth. We're excited to begin life as a public Company and to deliver long-term value to shareholders, employees, and other stakeholders.

Now, moving onto our operating results for the fourth quarter and 2022. We continued our strong growth momentum in fourth quarter. In September, the first Company in the world to deliver more than 10,000 LiDAR units in a month.

Total shipments during the fourth quarter reached 47,515 units, among which ADAS LiDAR shipments accounted for 91%. Note that in December, we shipped more than 20,000 LiDAR units. These achievements are groundbreaking in the industry.

In the fourth quarter, our new ADAS LiDAR product continued to win new customers, including, firstly, the largest EV maker in China, and secondly, the largest OEM, and thirdly, electric technology company, Rox, followed in January by Seres, a leading China-based E.V. OEM. These were followed by a design win with Didi's autonomous driving business. Additionally, in February 2023, the Company secured a LiDAR design win for AT128 with the Li Auto on its new battery electric vehicle platform.

Our successful Q4 capped a successful 2022. During the year, we proved that our mass production capabilities empower us to capitalize the realm of ADAS and autonomous mobility. Our total shipment in 2022 reached more than 80,000 LiDAR units and accumulated and surpassed 100,000 shipped units.

That represents an explosive 337% CAGR since 2020, significantly outpacing the global LiDAR market. In fact, our LiDAR deliveries and their revenues for 2022 were higher than the cumulative LiDAR unit deliveries and its revenues of all eight of our U.S. public listed peers combined in 2022. Notably, we shipped more than two times the number of automotive LiDAR units that our eight U.S. listed peers combined.

The backdrop of our success is our people. In 2022, we grew our talent force to more than 1,000, most of whom are on the R&D, actually in engineering team, a scale significantly higher than our peers. Our success is further defined by our innovative customers' and partners' in-house manufacturing capabilities and breakthrough LiDAR technologies.

In the ADAS market, we're working with many market-leading OEM partners, including Li Auto, Changan Automobile, Lotus, JiDU, HiPhi, Rox in autonomous mobility market where we are dominant.

Our customer base includes some of the most prominent global players, including world's leading robotaxi technology company in the U.S. Our customers also include Aurora, Pony.ai, WeRide, and [Algolux].

In the robotics market, we provide products for the largest player in China, Meituan, alongside global players such as Nuro and others. Our commitment to building superior products, we chose to optimize performance, quality and cost is a key to our wins in both ADAS and autonomous mobility markets.

We believe an advance in-house manufacturing capabilities are critical for developing LiDAR products like ours that utilize fast and advancing technologies. We combine R&D advances in application-specific integrated circuit, also known as ASIC, to drive semiconductor performance with in-house manufacturing capabilities to design, iterate, and bring the best-in-class products to our customers.

Our integrated in-house model enables fast development and proprietary knowhow to reinforce each other, establishing a virtual cycle and a formidable competitive advantage. Nuro also helps us to optimize cost, deliver higher performance solutions, and make our supply chain safer.

Through this model and our talented R&D and manufacturing engineering teams copy approximately 700 [indiscernible]. We developed our FT120 solid-state blind spot LiDAR sensor with no moving parts inside, designed for ADAS series production vehicles.

We plan to launch the FT120 later this year. In addition to FT120, during the course of the year, we expect to announce several really exciting industry-leading LiDAR products, so you have to stay tuned for those future release announcements.

We're proud that Hesai has become the most commercially successful LiDAR Company in the world. To-date, we signed 11 OEMs, six of which will be in mass production, and all 11 by 2024.

We're presently in discussion with several other leading Chinese and global OEMs, which we - which we expect to add in 2023 and look forward to sharing more information with you in due course.

Moving forward, we'll continue to build partners and relationships with OEMs and autonomous technology companies, perfect core LiDAR technologies and manufacturing capabilities. Our growth has just begun, and our path is clear. We're optimistic about the future trajectory.

With that, I'd now like to turn the call over to Louis, who will share about our financial performance for Q4 2022 and our outlook for the rest of the 2023. Louis, please go ahead.

Louis Hsieh

All right. Thank you, David. First, I want to thank everyone for attending Hesai inaugural earnings call as a public Company in its operating figures. Our commercial success is evident in our financial performance.

To be - to be mindful of the length of our earnings call today, I encourage listeners to refer to the earnings press release for the 2022 and full year results for further details. Clearly, Hesai had a stellar 2022 as we extended our market leadership in LiDAR solutions, both for autonomous mobility and now for ADAS.

Today, I will just go over the highlights of our Q4 and full year 2022 as I'd like to spend the bulk of our time together talking about our future. We achieved record net revenues of RMB409.2 million, equivalent to USD59.3 million in fourth quarter of 2022, representing an increase of 56.6% from the same period of 2021, and an increase of 22.6% from the third quarter of 2022.

Gross margin was 30% for the fourth quarter of 2022 compared to 52.4% for the same period of 2021 and 0.1% for the third quarter of 2022. Net revenues were RMB1.203 billion, equal to USD174.4 million for the full year of 2022, representing an increase of 66.9% from the previous year.

Gross margin was 39.2% for the full year of 2022, compared with 53% for the prior year. Many of you have asked about our gross margin profile as we enter the nascent but rapidly growing ADAS market.

2023 is a transition year for Hesai as we migrate from traditionally higher margin autonomous mobility of Pandar, XT and QT to the exploding but relative structurally lower margins ADAS sales of AT and FT.

To give you an understanding of the difference, our average ASP for autonomous mobility has been over USD5,000, with approximately 50% gross margin compared to that for ADAS, where the ASP is expected to be around USD500 digit to low double-digit gross margins by yearend.

In 2022, our ADAS AT sales accounted for about 25%, that's USD46 million, but expected to -account for 40% to 45% of revenue in 2023 with an expected more than 3x increase from the 62,000 units in 2022.

The bulk of the ADAS shipments will be in second half of 2023, as we expect to be SOP with all - at least OEM partners by Q3 2023. It will take some time for us to gain economies of scale in manufacturing, efficiency, and utilization, material procurement and labor intensity to reach our long-term target of 25% to 30% gross margins in ADAS for AT and its [indiscernible].

Against this very strong demand and order backlog backdrop, entering the first half of 2023, we are going to expand our manufacturing footprint with investment in two plants and to get both operational by Q2 of this year.

Let me give you more color about this capacity expansion. First, we are excited to confirm that our new in-house manufacturing facility, Maxwell, a state-of-the-art comprehensive art innovation design center, manufacturing and testing complex in Shanghai, covering over 600,000 square feet will come online in Q2 of this year.

Maxwell will expand our manufacturing cap of 1 million units in terms of annual shipments and will equip us with a full-range testing capability. Second, in late Q2 or Q3, we plan to open and start another AT production line in our 300,000 square foot new manufacturing facility in [Hong Kong], which will also eventually have capacity for over 1 million units annually and be highly automated.

Along with our two smaller production facilities, which currently manufacture Pandar, QT, XT, and one line for AT128, our [total] capacity will exceed 2 million units on an annual basis by yearend.

The combined effects of product mix with ADAS increasing in revenue mix from approximately 25% to 45% of revenue and the opening of two large production facilities which will take time to reach optimal scale efficiency and utilization will put downward pressure on blended gross margin percentage in 2023, but rebound somewhat in 2024. Long-term target for blended gross margin remains at 33% to 35% for Hesai.

Now, I'd like to turn to our business outlook. For the first quarter of 2023, the Company expects net revenue to be between RMB390 million, USD56.5 million, and RMB410 million, USD59.4 million, representing a year-over-year increase of approximately 57% to 65%.

The above outlook is based on the current market conditions and reflects the Company's preliminary estimates of market and operating conditions and customer demand, which all are subject to change.

This concludes our prepared remarks for today. Operator, we are now ready to take questions.

Question-and-Answer Session

Operator

[Operator Instructions] Your first question comes from Olivia Xu from Goldman Sachs. Please go ahead.

Olivia Xu

Hi, David and Louis. Congrats on the Company solid performance. This is Olivia Xu from Goldman. May I have two questions? The first one is about the partnership. We noticed that Hesai has made a breakthrough in the partnership expansion and product nomination, including BYD, Seres and then the [Auto Pure Easy] platform.

On the Li Auto side, the Company has recently launched the L version of L7 and L8. Just wondering where it cannibalize the high end volume that impacts the LiDAR installation on the two models.

And in addition, is there any more information can be shared on the supply relationship of - to BYD and the Seres, for example, the timeline, which car models and the estimated volume? Thank you. That's my first question.

David Li

Hi, this is David Li. Thank you for the question. Let me maybe address the Li Auto one first because it's one of our largest clients. So, you're right that the LiDARs are - or the standard configuration for the L9, and then they are optional for the L8 and L7.

And for both editions, they have the AD Max, which is the one with LiDAR, and the - and AD Pro, which is the one without LiDAR. But having said that, the L7 and L8 are larger volume products so - and the take rate is actually pretty decent even today. And that's why we're able to come up with our projections for the rest of the year based on the take rate of the L7, L8 and the L9 and all the LiDARs, not from all our clients.

And also, one interesting note is that - and if you watch very carefully of the Li Auto L7 release, they actually mentioned that now, by end of the year, they're going to have more urban NOA, Navigation on AutoPilot, functions when it's with LiDAR, which means that the version with LiDAR will be much more valuable in terms of the - its ability to provide advanced driving functions.

And that was actually a very big boost on the take rate of the Max version of the L7, so which is extremely encouraging to know that even L7 is technically considered a more affordable version. You would assume that people wouldn't have a higher take rate for the one - the configuration with the LiDAR and turned out that it's not entirely true. People still like it, people still buy that with the more expensive configuration because they really look forward to the driving function that will be available down the road.

And also, on top of it, I want to point out the fact that when you buy a smart E.V. today, you're paying the price today, but you're not getting the full function yet. Most of - a lot of the functions, especially on the autonomous driving side, are being released via OTA, over-the-air to you over time, which means that technically, you pay the money already, but the value will go up over time, which is a good signal for us because that's why we believe over time, more and more people will be buying into the function as the value go up. That's why long term, we're very optimistic on that.

And now, and do you want me to comment on the Seres and BYD, right? It's confirmed that we started working with some of the models. Unfortunately, we don't have - I don't have more information about the more models that we're in discussion with them and some of them have not come to the decisions yet.

But what I could say is that because both Seres and the BYD is not on their entire fleet yet. There are quite a few exciting design wins that could come out later this year, or even some of them are very soon. So, we look forward to it and we'll keep you updated on that.

Louis Hsieh

Olivia, this is Louis. For Seres, we do expect to ship this year. BYD is more likely - the volume will be next year when their models come out.

David Li

Yes.

Olivia Xu

Thank you. That's my first.

Louis Hsieh

Does that answer your question?

Olivia Xu

Yes, right, that's very helpful. My next question is about the implication of the recent E.V. pricing cut. We noticed that year-to-date, Tesla has a regular price in China and a lot of other E.V. makers have to revise their pricing as well.

Given the diminishing government subsidies, it is likely that the carmakers' margin will get squeezed. Just wondering, does the pricing pressure has passed to the LiDAR suppliers now and was that a responsive strategy towards the E.V. pricing cut? Thanks.

David Li

Yes, we are observing that now also already. What I like to point out is that if you look at the pricing cut phenomena, it's really on the more affordable part of the spectrum of the market.

If you think about it, the nature of LiDAR, it's at least now and is - the penetration is mostly on the more premium part of the market, especially if you look at Li Auto. I don't think they have sold anything that's below RMB300,000.

And it's at least from what we received and observed that it is much less impacted, if at all by the pricing cut. And we do expect a lot of competition on the below 200K level for the cars, but those are also not our typical choices for cars when they want to buy LiDAR.

If you're buying a car that's, like, costing you RMB150K, you probably don't really care about LiDARs today, right? You want the LiDARs to get on all the big screens and all that today, the batteries, right?

And it only becomes a much more serious interest when you're paying for, I would say, at least RMB200,000, ideally RMB250,000. And that's the range where people started to look for things that's beyond battery and infotainment system, and that's where LiDAR became a symbol of intelligent driving.

Louis Hsieh

I think, Olivia, for us, as David said, we're basically more in the higher end of each of these lines, so we probably wouldn't see as much impact from price reductions at the lower end of the - of their pipelines. But for us all, six OEM shipping this year, the majority will come in in Q3 and Q4. They will be in cars above RMB250,000. So, they're less impacted by price type competition.

And then the bulk of the volumes will be in '24. We expect volume to at least double or triple from 2023 levels. And that's when we'll, I guess, see if the price increases will affect the demand.

David Li

Okay.

Olivia Xu

Okay, got it. That's very clear. Thank you, David. Thank you, Louis.

David Li

Thank you, Olivia.

Louis Hsieh

Next, please.

Operator

Thank you. Your next question comes from Tim Hsiao from Morgan Stanley. Please go ahead.

Tim Hsiao

Hi, David. Hi, Louis. Congratulations on the solid result. I've got two questions. So, the first one, I think David already touched on some feedback and observations about the consumers on the LiDAR adoption.

But based on your recent conversations with carmakers, is there any change - any change to their thoughts about adoption pace of LiDAR during the mounting pricing pressure [being on the question]?

Will there be any risk of near-term downgrading to expect or of their upcoming models with lower adoption of LiDAR or actually the carmakers would be getting more aggressive to upgrade the spec with more LiDAR adoptions to differentiate themselves? So, this year, it was a little bit about what the feedback on the carmaker may be? Thank you.

Louis Hsieh

Yes. Tim is your question about, in our conversations with carmakers, are they more or less inclined to use LiDAR for next year sort of what you're getting at?

Tim Hsiao

Yes, yes, because I think some of the carmakers might consider to be more cost-conscious and buy more, in fact...

Louis Hsieh

Yes.

Tim Hsiao

...[indiscernible] but probably the other customers they're even more aggressive in considering to adapt LiDAR. So what's the feedback of your major customers because of that?

Louis Hsieh

I think this goes in line with Olivia's question is that our LiDAR are typically on the premium models, so less likely that they will cut this because it is a marketing play, right? So if you have a LiDAR and your competitor doesn't have LiDAR, it's actually a big selling point.

In addition, these models are coming on '24, so what they don't want to do is their new generation models be seen as really standard models with no differentiation. So, so far, as far as I know, we have not seen carmakers in our 11 OEMs, they say they're going to cut LiDAR out of any models or reduce the effect of LiDAR in those models.

Those models are on track, one thing that may happen is if there's a delay in the SOP of some of these models in '24 that several of these carmakers, these are brand new models, so there is some delay risk on that side. But I haven't seen anywhere they actually cut the LiDAR out.

Have you, David?

David Li

Correct, yes, no.

Louis Hsieh

Did that answer your question, Tim?

Tim Hsiao

[Indiscernible], yes. Thank you, Louis. My second question is about the manufacturing ability because I think this year's important milestone would be the launch of our new plant, Maxwell. So would you please share with us the current progress of our new plant? Is it still on track to kickstart the next production in the midyear and all the ramp-up [indiscernible] on track to meet our target? Thank you.

Louis Hsieh

Okay, okay. We have Sun Kai, one of our other co-founders here and Chief Scientist. He can answer the question. When will Maxwell be operational?

Kai Sun

Okay. So for the [indiscernible] will be finished in Q2 this year, probably the middle or late Q2. And we'll have some automatic line by the end of this year, yes. That's the plan for Maxwell.

Louis Hsieh

And then for Hangzhou, Tim, that will come after Maxwell. And we expect to have one or two operational lines in Hangzhou. This will be for new redesigns. So the existing AT128, we have one line already in Shanghai that's producing.

As we said in the earnings release, our goal this year is to work on the gross margin profile since ADAS typically has a lower gross margin profile. So we have a redesign of the AT coming out, and we don't want to put it in the same manufacturing line as the current AT. So the Hangzhou plant will be most likely the current plan is they use one line for the new AT redesign that will be much lower cost and that will improve the gross margin profile.

Okay. You have to stay tuned [indiscernible].

Tim Hsiao

Thank you very much [indiscernible].

Louis Hsieh

Okay. So that's why we opened two [indiscernible]. We don't want to cross the two production lines, [indiscernible] one and then one manufacturing a later model the same one. Plus, we believe we will use that capacity up by '24, '25. So we're just getting the factories ready ahead of the expected demand, which we think is a prudent thing to do. Okay.

Tim Hsiao

Okay, thank you, Louis.

Louis Hsieh

Thank you, Tim. Next question please.

Operator

Thank you. Your next question comes from Bin Wang from Credit Suisse. Please go ahead.

Bin Wang

My question [indiscernible] global automaker [indsicernibe]. If any chance this year we can [indiscernible] from the global automaker [indiscernible]? That's number one question.

Number two is [indiscernible] margin. [indiscernible] last year there's been about 30% gross margin. Can you provide breakdown between the ADAS, LiDAR and the [indiscernible]. And what's your [indiscernible] first quarter 2023 and the full-year for the two LiDAR products? Thank you.

Louis Hsieh

David will take the first question.

David Li

Yes.

Louis Hsieh

And I'll take the second question.

David Li

Yes. Thank you, Louis. So I guess, I'll take that. Should I be doing it in English or Chinese?

Louis Hsieh

English.

David Li

Okay, cool. So question is not about the global side of the business. And it's hard for me to directly comment the competitors because at their right to disclosing, but I want to point out that in our definition, when we say a global deal and we consider only from the global programs, right, and one is talking about the localization of some of the joint venture car models and all that. And we actually have some of them, but for us it's a China deal. For global, truly global, it has to be sourced from the car models that are shipping globally, which are typically the European and Americans.

And so, we are in late-stage discussions with multiple, including Europe and American. That will come out in Q3 and Q4 as the final results this year. We're in the either RFI or RFQ phase of them. So we are very optimistic about it for the reasons, I think, we explained in the past. I'll quickly repeat them.

One is that this round of sourcing from the global OEMs, they're really focused on the practicality of stability to deliver. And in the past., a lot of the global peers have not demonstrated the capability to deliver volume products as ADAS.

As you can imagine, this is one of the most critical ability for any automotive vendors to do. And we illustrated that. We shipped more than 80,000 units in total. And I think the official number for the ADAS sensors, that's close to 50 ...

Louis Hsieh

[Indiscernible] 62,000.

David Li

... 60,000 ...

Louis Hsieh

Yes.

David Li

... ADAS sensors. And that's one of the numbers that OEM really care about today because they want to make sure you have the ability to deliver. And the one evidence you can use is product [ph] shipment, and now we have it. And the other one is actually cost.

And historically, for the first round of the sourcing from a few years ago from the global players, they were the smaller volume vehicles. And for that, it's probably okay it's at a relatively higher price. Now, most of them are being much more serious about rolling those configurations into more affordable price range large-volume vehicles and in which price becomes a much more critical issue.

And for us, with our ability and history of developing in-house [indiscernible] and in-house manufacturing, we have the best ability to do low-cost ADAS sensor at a reasonable margin. That's what they care about. That's why we have their high confidence that at least some of the deals will come to us. And hopefully, we'll be announcing that.

Louis Hsieh

Right. Thank you, David. Great questions, Wan Bing.

On the gross margin side, in 2022, our gross margins for the autonomous mobility side was over 50%. On the ADAS side, it was less than 5% because we were just ramping AT. It was only for several months. So remember the production line of the AT128 that's currently in effect has capacity for 300,000 units a year. We only manufactured 60,000 and delivered in 2022.

This year for Q1 and Q2, utilization rate will not be quite as high. By Q3-Q4, it will ramp up, so we expect to ship at least 200,000 AT plus a few units this year. So it will get better with the gross margin profile.

So for this year, on AT, it will go up from under 5%. By the end of this year, we expect the gross margin to be in the high single-digits to low double-digits, so probably 9% to 13% as we exit this year. The long-term gross margin we expect for AT should be around 30% to a little bit over 35% as we redesign the model and as we go to mass production. Remember, in '24, we have 11 OEMs shipping in AT, so the volumes will be just that we can get the higher gross margins.

On the autonomous mobility side for 2023, this year, we are undergoing in Q2 and Q3 a price reduction on the blended basis for Pandar. So the ASP last year was about 13,000 for autonomous mobility products. This year we expect that number to be around 9,000. So because of the price cut, it will put some downward pressure on the gross margin profile for probably Q2 and Q3 of this year. So Q1 should be okay. But Q2-Q3, it will have some negative impact.

And so, I wanted to also say that we are addressing this as well. So we will have a lower cost Pandar model coming out in the second half of this year. That will significantly reduce our production cost, and we'll still be able to maintain our margin at around 50% for this product at scale. So that will mean by the end of next year, the margin will bounce back.

So as you asked earlier, our long-term gross margin target remains unchanged at 35% where we have autonomous mobility at between 45% and 50%, and ADAS at probably 2027 to 33% you're with the target 30% or so. That blended number should come out to 35%, so nothing has changed on the long-term perspective. But on the transition for this year, the first half will be slower. It's only the seasonal slower time for us, Q1-Q2. In Q3-Q4, the margins will begin to ramp up again.

Does that answer your question, Wan Bing?

Bin Wang

And, sir, can I have a follow-up? I think [indiscernible] quite decent in the [indiscernible] for the first quarter. Can I assume the possibility to secure the [indiscernible] 90% of [indiscernible]?

Louis Hsieh

Sorry, I didn't hear that, repeat. Yes, repeat.

Bin Wang

I mean, you already give the timing for the quarter, number four quarter, potential [indiscernible]. Can I assume there's a possibility to announce a global ADAS by the customer [indiscernible] high?

Louis Hsieh

[Indiscernible] getting global ADAS announcement in the first half of...

David Li

First half?

Louis Hsieh

Did you mean first half of this year, Wang Bin?

Bin Wang

No, no, no. I mean, yes or no. Whether this year you will announce [indiscernible] from global automakers [indiscernible] possibility is directly from [indiscernible].

David Li

For this year, we have actually multiple European and at least the one American major design decisions coming out by Q3 and Q4. So the chance of getting more than one, at least one, is very high.

Bin Wang

Okay, thank you.

Louis Hsieh

But then also, Wang Bin, I will remind our investors that the adoption for China is actually much more rapid than the U.S., right? We have 11 Chinese OEMs. The volume in China will be four or five times the volume for the U.S. and Europe in the next two or three years.

So we went after where the market is, and the market is China today. So we went after and we have 11. We have a dominant share in China already.

On the global side, the game is just beginning in '23 for the volume production in '26'-27. So we are fighting and we are in the game, but the results won't come out until second half of this year. Okay.

David Li

And also the actual shipment will only...

Louis Hsieh

Start in ‘26 and ‘27.

David Li

Yes, as early as '26.

Louis Hsieh

Yes. So for the next two years, it's more important to focus on our China deliveries because that's where the revenue drivers and that's where the knowhow and the production will come from. And that's where the margin will come from. So that's why you can see. The reason is we completely dominate our global peers is because they have no volume in production. So we're larger than all eight combined because we are winning China, where that's where the ball game is. You could only play where there's a game. And the game for North America and Europe really won't be -- you start scoring in '26 and '27. Okay.

Louis Hsieh

Thank you, Wan Bing. Next question please.

Operator

[Operator Instructions] Your next question comes from Paul Gong from UBS. Please go ahead.

Paul Gong

Yes, hi. Thanks, management, for taking my questions. I have two questions. The first one is you mentioned that the mechanical LiDAR has more than 50% gross margin, while the ADAS LiDAR has less than 5% gross margin in 2022. So from the revenue of the Q4, is that fair to assume the revenue mix or revenue split is roughly like 50-50 between the robotaxi LiDAR and ADAS LiDAR? Is that a fair estimate of the revenue split?

Louis Hsieh

For Q4, that's pretty close. For the whole year, ADAS is 25% and autonomous mobility was the other 75%. So going forward, which is now in '23, ADAS should be 40% to 45% of revenue in this year. But as I said earlier, the bulk of the units will come in Q3 and Q4 as new models are released. So that's the seasonal factor of our business is that Q3-Q4 typically are much larger quarters than Q1-Q2. So Q1, the reason the gross margin will be better is because there are more autonomous mobility shipments in Q1. Okay.

David Li

And the other quick point I want to make on the gross margin side is it's true that last year the ADAS gross margin wasn't as we like to be but remember that this is the first time we're shipping a larger volume ADAS, right? And then if you compare that number to all the peers, that's still not crazily bad considering most of the people are having really minus gross margins.

Louis Hsieh

And, Paul, I'll give you are stats so you understand how fast we are working on the cost side. The average COGS for AT in 2022 was $700. The average COGS today is under $500, okay.

So as ASP come down because we have a contract, the cost is also coming down commensurately, if not faster, but there's a transition period, right? It's another quarter or two to ramp-up the new production line in Hangzhou for the new AT. That has significant price reduction versus the current version that is in Shanghai. And that's why there's a slight lag in the gross margin profile as a new line comes up, and then it needs to scale.

But I think this is the right decision for us. We invest this year, and '24 should be a blockbuster year for us in terms of the deliveries, right? I mean, we expect to be delivering 600,000 to 800,000, even maybe close to 1 million units in '24 when our competitors aren't even at 50,000 yet. So we're basically trying to extend our lead and speed of manufacturing.

Paul Gong

Understood, understood. So just to double-check, you said for the full year this year, the ADAS LiDAR is going to contribute roughly 40% to 45% on you. Is that right?

Louis Hsieh

Correct, correct, about $115 million to $120 million, yes.

Paul Gong

Okay. And the long-term margin for the robotaxi LiDAR is 45% to 50%, and target for the ADAS LiDAR margin is 25% to 30%, right. These are the right numbers to calculate?

Louis Hsieh

Yes. That's about correct. Yes, that’s correct.

Paul Gong

Okay, okay.

Louis Hsieh

We’ve blended out, because ADAS is growing faster. It'll be [35%] target.

Paul Gong

Understood, understood. My second question is regarding the Q1 guidance. I understand right now the ADAS LiDAR [indiscernible]. I think the auto [indiscernible] is somewhere like 10% to 20% quarter-over-quarter [indiscernible] growth in Q1 versus Q4 or even though maybe the mix is going to be more skewed towards L8 instead of L9. So that basically means the Q1 versus Q4, your guidance is also flattish. That basically means flattish in both the ADAS LiDAR, as well as the robotaxi LiDAR. That's the right way to understand, yes.

Louis Hsieh

Yes. To be honest, that's not entirely correct for us. So you didn't know, in Q4, we have higher pricing for ADAS LiDAR, so we ship more units. So that's why we shipped 20,000 units in December. So Q1, there'll be a slight slowdown in ADAS for January/February as those December units are used up. In March, we'll pick up again.

So I would expect my current forecast is only about 30,000 ADAS units in Q1. Robotaxi will have a very strong Q1. And that's why you see the guidance where it is.

You say it's flattish over Q4, but you have to remember Q1 is seasonally our slowest quarter because Chinese New Year, and also there's usually a lag effect. Q4 is the biggest quarter. Q1 and Q2 are slower. So for us to be flattish over Q4 is quite an accomplishment. It's over 60% year-over-year growth.

And, Paul, you've known me for four years, five. You know, I guide conservatively.

Paul Gong

Understood, understood. And ...

Louis Hsieh

So it could be the first time, it could be we will have a record quarter, the biggest quarter of our history.

Paul Gong

Understand. So do you feel that the mix is going to be most [indiscernible] towards the robotaxi or LiDAR in Q1. That also means that ...

Louis Hsieh

Only in Q1, yes.

Paul Gong

Yes, that also means the margin profile would be even better than Q4.

Louis Hsieh

Yes, and that's what I told you earlier. I said that already.

Paul Gong

Okay.

Louis Hsieh

So [indiscernible] Q2, but in Q2 ADAS takes over. And Q2 is when a price reduction in the robotaxi LiDAR takes effect. So Q2 then will have a negative impact on the gross margin because the highest gross margin product is taking a price cut before the new design goes into effect in Q4. So there's a two-quarter impact on the gross margin for the robotaxi because of a price decrease where the reduced price model, cost model won't come in until Q4. So that's why I'm letting you guys know ahead of time so you properly model this. It's a short-term blip, but it's important that we don't mislead you.

Paul Gong

Yes, that's very responsible and very transparent guidance. Thank you so much for the color and...

Louis Hsieh

So that, you know me, I tell you straight, good or bad.

Paul Gong

Okay, okay. Thank you so much.

David Li

Thank you, Paul.

Operator

[Operator Instructions] As there are no further questions now, I would like to turn the call back over to the company for closing remarks.

Rachel Yang

Okay. Thank you once again for joining us today. If you have further questions, please feel free to contact our Investor Relations through the contact information provided [indiscernible] financial communication. Thank you.

Operator

Thank you. This concludes today's conference call. You may now disconnect your line. Thank you.