Home / Transcripts / WeRide Inc. (WRD) · May 13, 2026

WeRide Inc. (WRD) Earnings Call Transcript

May 13, 2026

NASDAQ US Consumer Discretionary Automobile Components earnings 62 min

Earnings Call Speaker Segments

Xu Han executive
#1

Hi, everybody, and thanks for being with us today. As a global frontrunner in autonomous driving, we entered 2026 with a lot of momentum. During Q1, WeRide's overall revenue came to RMB 114 million, which marks a 58% increase compared with the same period a year earlier. Those numbers come from our speeding up global taxi rollout, expansion throughout our wider autonomous driving operations, and the tremendous success of our L2++ level ADAS offering. As of the end of April, our worldwide robotaxi fleet had expanded to roughly 1,300 vehicles, making it one of the biggest taxi fleets of its kind anywhere. Meanwhile, our Level 4 autonomous driving fleet, which includes the robovan and the robobus, has reached approximately 2,800 units. Those — they've been deployed or tested across 12 countries and more than 40 cities around the world; we think the research and notable strides made this quarter speak not just to how mature our technology is, but also to the expanding operational know-how for < > level fleet in numerous cities. To begin with, I'd like to highlight that we achieved a significant technical leap with GENESIS, our closed-loop world model-based simulation engine, which sped up our model evolution pace several times over; we can now train AI models for autonomous driving using synthesized corner cases, which might be extremely rare or even purely imaginary. Using a compact AI model powered by our GENESIS, we've won 4 straight championships in China's urban intelligent driving competition; that's never happened before. The prior best record belongs to the Huawei ADAS system, which earned 2 consecutive championships. Nowadays, plenty of companies discuss world models and simulation platforms. Yet WeRide is among the very few that have publicly shown footage of a highly realistic autonomous driving world model at scale; we've already put a GENESIS demonstration video on YouTube, where viewers can examine how well the model reconstructs and simulates a visually realistic driving environment, while strictly adhering to physical laws as well. GENESIS is able to produce a holistic virtual driving environment matching the desired locations, covering traffic laws, pedestrians, weather conditions, and intricate interactions among vehicles and surrounding objects. Even more importantly, we can introduce this environment component to simulate highly challenging scenarios like aggressive driving behavior, bad traffic, extreme weather, our euro road conditions, and many other long-tail corner cases that are extremely hard and very costly to replicate in the real world. Through the use of synthetic data and large-scale simulation, GENESIS raised training and validation efficiency by thousands of times relative to traditional road testing. It also greatly cuts the critical reliance on a large-scale testing fleet and speeds up the deployment process in new operational regions globally. GENESIS isn't merely a capability; it's a unified simulation and AI training platform that supports applications ranging from L2 ADAS to L4 robotaxi. And it's that very same < > that makes us the only company worldwide to have already accomplished scaled commercialization of both L4 and mass production L2 vehicles. The newly built GENESIS has now opened the path for WeRide into the physical AI world. When you look at Genesis, the ADAS system that WeRide built is on par with what FSD 14.3 delivers in California. I own two Teslas myself and use FSD 14.3 daily. I'm eager for us to get into a worldwide urban intelligent driving contest that puts us straight up against Tesla's FSD 14.3, and for the day when we could meet in the U.S. or Europe and offer our customers a direct head-to-head comparison. Right now, our technology leadership is showing up as genuine commercial scale around the world. Let me take you — let me guide you through the important operational and commercialization milestones from this quarter. To start, in China, our robotaxi business kept delivering solid gains across scale, operational efficiency and depth of commercialization. To start, in China, our robotaxi business kept delivering solid gains across scale and operational efficiency. As of the close of April, our domestic global taxi fleet had grown to roughly 1,000 vehicles, and our Guangzhou service area grew 97% versus the close of 2025, adding more downtown districts. On the demand side, average daily orders per vehicle domestically hit 17 trips in Q1, with peaks reaching 28 trips per vehicle. The count of registered robotaxi users also nearly doubled annually. We think these metrics keep showing rising user adoption and better unit economics as robotaxi commercialization scales. During this quarter, we also kept deepening our partner ecosystem. In April, we broadened our work with Lenovo on autonomous driving computing platforms, with a shared goal of deploying 200,000 autonomous driving vehicles worldwide within the next 5 years. Alongside Geely Farizon, we intend to deliver 2,000 upgraded, purposely built robotaxi GXRs in 2026. We think this partnership further boosted our manufacturing scalability and our deployment capability globally; now shifting to international markets. We also keep seeing strong momentum both in launching new markets and in commercialization progress. In Singapore, we started the country's first public autonomous driving service alongside Grab. Since the initial development — I apologize, since the initial deployment plan started in the second half of last year, the fleet has established a trustworthy operational track record in a heavily regulated international market. In the Mid East, we began fully driverless commercial robotaxi operation in Dubai alongside Uber and Dubai's RTA; this marks the city's first fully driverless commercial robotaxi service Meanwhile, Abu Dhabi service coverage grew to about 70% of the city's core area. Throughout the Mid East. Working with Uber, we remain on track to deploy at least 1,200 robotaxi across Abu Dhabi, Dubai and Riyadh by 2027. In Europe, we entered Slovakia in March, our fourth European market, and kept moving toward fully driverless commercial operation in Zurich. All in all, we continue to believe WeRide remains the most globally deployed autonomous driving company today, with deployments across 12 countries and permits in 8 markets. This global footprint not only diversifies our revenue but also demonstrates operational and regulatory capabilities that are hard for followers to replicate. That additionally maintains our momentum toward the eventual goal we've set: getting tens of thousands of global taxis onto roads worldwide by 2030. Past Global Taxi, commercial interest keeps building in our ADAS operation as well. Nearly 30 vehicle models now incorporate WRD 3.0, the current edition of our ADAS system, among them cars from top OEMs like GAC and Chery. GAC Aion began presales during April for the Aion N60, the first mass-produced car featuring WeRide's technology, and thanks to GENESIS's generalization abilities, our WRD 3.0 offering now works across 3 major chip platforms: NVIDIA DRIVE, Qualcomm Snapdragon, and SiEngine StarLight. We see this multichip compatibility as essential, since it drives quicker mass production, better cost optimization, and wider OEM uptake. Concurrently, we keep growing our international footprint alongside partners, among them < >, delivering a WeRide powered ADAS solution to consumers around the world. Lastly, regarding robobus, we keep advancing worldwide too; alongside the Geneva public transport operator, TPG, we are working on our autonomous bus deployment. In parallel, we are getting ready for robobus operation with < > branch open for the third consecutive year. In summary, Q1 of 2026 centered on turning WeRide's technology leadership into commercial scalability, via expanding robotaxi operations, entry into additional international markets, a growing deployment pipeline, and continued unmatched winning momentum in ADAS. We think we're executing our global strategy effectively, and a clear route to long-term growth remains visible; with that, I'll hand the call to our CFO, Jennifer Li, who will give a deeper look at this quarter's financial results. Thank you.

Xuan Li executive
#2

Thanks, Tony. Hi, everybody. Ahead of getting into the numbers, I'd like to point out that every figure is denominated in RMB, and all comparisons are made against the prior year, unless we say otherwise. We'll also be covering non-IFRS metrics today, and those are explained more fully, and reconciled to the closest comparable items disclosed in the company's earnings release and its filings with the SEC and the Hong Kong Stock Exchange. Let's now turn to how we performed financially in the first quarter; total revenue came in at RMB 114 million for Q1, which works out to growth of 38%. Product revenue rose 116%, reaching RMB 20 million, largely on the back of greater deployment of robotaxi and other L4 vehicles. Service revenue climbed 49%, hitting RMB 94 million. This top line expansion reflects the strong commercial strides we took during the quarter, along with our demonstrated ability to execute and deploy. Given the seasonal effects of Chinese New Year and Ramadan in the Middle East, our results came in ahead of the internal targets we had originally set. Gross profit at the group level rose 56%, reaching RMB 40 million in Q1, and our group level gross margin stood at roughly 35%. We kept growing the top line while holding the line on margin discipline, which showcases the built-in profitability of our autonomous driving business as we keep scaling. Additional support came from greater exposure to our ex-China market, where we keep observing a structurally stronger margin profile as we move into further new international territories. Operating expense came to RMB 459 million, with R&D expense making up 77% of that total. Our operating expense held steady in absolute dollar terms relative to the same period in 2025. Breaking it down further, R&D expense rose by 12%, reaching RMB 353 million in Q1 2026. If we set aside share-based compensation, R&D expense was up 16%, at RMB 322 million. This steady investment in R&D supports our technology road map and keeps us at the forefront of autonomous driving innovation. Administrative expense fell by 33%, to RMB 83 million in the first quarter. Setting aside share-based compensation, administrative expense declined by 17%, to RMB 51 million. That drop came mostly from lower professional services fees, primarily tied to audit and legal compliance services, and was partly offset by higher personnel costs from expanding our team as the business grows. Selling expense rose by 63%, to RMB 23 million in Q1 2026. Setting aside share-based compensation, selling expense was up 81%, to RMB 22 million. That rise aligns with the ongoing expansion of our business and demonstrates our commitment to supporting growth in an appropriate way. Our net loss came in at RMB 359 million for the first quarter. On a non-IFRS basis, net loss grew modestly, by 11%, to RMB 326 million in the first quarter. That slight increase stemmed largely from continued R&D spending and from our ongoing practice of investing ahead of scale in our long-term technology leadership. By the close of March 31, 2026, our total capital reserve stood at RMB 6.22 billion. That figure breaks down into RMB 6.18 billion held as cash plus cash equivalents and time deposits, RMB 29 million sitting in investment and wealth management products, and RMB 18 million of restricted cash. To keep day-to-day operations running, we carried short-term bank loans totaling RMB 294 million. Our capital base is now well positioned, aligning what we need for cash deployment while leaving headroom for continued growth and strategic initiatives. On March 23, 2026, our Board of Directors authorized a share repurchase program of USD 100 million. Through market close on May 12, we had bought back roughly 24.4 million Class A ordinary shares, some of them in the form of American depositary shares, for total consideration of about USD 61.4 million. That speaks to our strong belief in this company's long-term value and growth potential. Looking ahead, we move forward with confidence and a clearly defined focus. By the end of 2026, we expect to stay on track to put 2,000 robotaxis into operation globally — the first milestone on our path toward hundreds of thousands by 2030. As we keep scaling, our strategy is to expand into new regions and studies that have demonstrated effectiveness and replicability. Backed by strong technological leadership, operational know-how, and an increasingly robust global rollout, we are ready to lead the next phase of autonomous driving. With that, operator, we are now ready to take some questions.

Analyst analyst
#3

This is Stanley calling from Morgan Stanley, and I've got a couple of questions for you. The first relates to how your robotaxi rollout is going — will you hit the road map you set out for 2026? Also, could you bring us up to speed on where things currently stand with expansion in China, the Middle East, and elsewhere around the globe? My second question concerns WeRide's business abroad. Your company's main edge lies in holding overseas operating licenses across 8 countries, and given how players such as Waymo and Zoox are pushing globally, plus regulators granting approvals at a faster pace, how do you see the main growth drivers, the route to profitability, and the timeline for overseas operations delivering meaningful contributions between 2026 and 2027?

Xu Han executive
#4

Alright, I'll handle the first one, and then Jennifer can take the second one regarding profitability, okay? So the first one asks whether we're on track with robotaxi expansion. Broadly speaking, we feel very optimistic and confident about growing our robotaxi business worldwide, and progress has been steady throughout. At the — let me give some figures already shared with the market, which is — by the close of April, our robotaxi fleet around the world stood at roughly 1,300 vehicles. As far as I know, that ranks among the largest robotaxi fleets anywhere. Within China, that fleet has reached about 1,000 vehicles, backed by solid demand, healthy daily order volume, and a user base that keeps growing. One figure I'd like to pass along: during the first quarter, average daily orders per vehicle came in above 17 trips, which is a remarkable figure. Meanwhile, our robotaxi fleet overseas has grown to roughly 300 vehicles spanning multiple markets; in the Middle East, we recently began full driverless commercial operation in Dubai and on the Uber platform. Right now, if you truly want a full driver, it's robotaxi. From what I know, looking at our competitors, outside the U.S. and China, the sole route to hiring a fully driverless robotaxi runs through Uber hiding WeRide's robotaxi in Abu Dhabi or Dubai, okay? That's where we stand at present. Regional tensions have created — I know regional tension brings some near-term softness in utilization. Even so, we stay very confident. More importantly, we stay firmly committed to the Middle East, and local governments have welcomed our long-term investment and operational presents. We stay on track toward our committed commitments to deploy 1,200 vehicles across Dubai, Abu Dhabi, Riyadh as additional driver is ODD to be added; in Europe, our footprint keeps growing. In Switzerland, we secured regional's first driverless commercial permit and keep progressing toward public operations in Zurich; we also recently kicked off our national autonomous driving program in Slovakia, marking our entry into yet another new European market. More broadly, we believe WeRide has built a meaningful advantage, first-mover advantage through years of global expansion. And on the whole, I'd say that having operations and regulatory reach around the world is turning into a bigger and bigger advantage as self-driving commercialization speeds up across the globe. So with that, I've finished addressing what you asked. Our worldwide road map and our plan, alright? On the topic of profitability, Jennifer, would you take that question?

Xuan Li executive
#5

Sure, I'll take the second question. Yeah. So, as Stanley pointed out, one benefit is holding operating licenses across 8 countries. In fact, if you look at how many robotaxis have been put into service and how many driverless permits are held by any autonomous driving firm outside of China and the U.S., WeRide comes in first among — across all those different businesses, 4 businesses; the momentum behind our global growth so far gives us a lot of encouragement. As we noted in the prior earnings release, revenue from abroad made up roughly 1/3 of overall group revenue during last year. Our Middle East subsidiary had already reached net profitability; and — as Tony just covered, he already went through our major developments in certain Chinese regions this year, and we'll be sharing additional exciting news about global taxi deployment when the time is right. So for this year, we anticipate revenue from abroad rising at a quicker pace and making up a bigger portion of group revenue, backed by favorable economics. Perhaps I'd also like to say a bit more regarding the way we pick cities. People often ask us, you already operate in something like 40 cities, so how — are you looking to add more cities or concentrate on the ones you have; our approach is to pursue scale in cities where we already are while also moving into new ones, and the criteria are more selective. We genuinely have to — we genuinely weigh whether a market can sustain scalable, commercially appealing operations over the long run. Particularly when it comes to monetization potential, we watch total gross booking opportunities closely, along with gross bookings per mile, since both scale and unit price play a major role in long-term global taxi unit economics; on top of that, we place strong emphasis on whether a realistic route to scale exists for every city where we're currently deploying; we do see potential for every city to deploy thousands of autonomous driving vehicles, yes. One more thing worth noting, Europe stays a key priority for us this year. We hope to share our exciting news about more deployment in additional European cities before long. Yeah, that wraps up my response to this question.

Ming-Hsun Lee analyst
#6

There are a couple of things I'd like to ask. First, in what ways does the latest report regarding China's pause on fresh self-driving approvals affect WeRide? Then my second one concerns differences in technology. For a long time, I believe, there's been an ongoing debate across the industry about the LIDAR plus HD Map strategy Waymo uses versus Tesla's camera-only route — where does WeRide stand?

Xu Han executive
#7

Let me handle these sequentially, all right? I jotted some things down. Now, regarding your first point. I think everyone is already aware of the recent buzz surrounding the latest suspension of new self-driving approvals; all of us noticed that sort of accident involving Baidu in Wuhan. We've gone through those reports. To start, how we see it is that this looks like a short-term tweak from regulators instead of a fundamental shift for the sector, since we've been talking with both the central and local governments, and through that dialogue along with the thorough probe — first off, if you look at our track record, WeRide's safety and operational history is very strong, and the central government as well as local governments have given us really robust backing. From our conversations, we came away confident that backing for self-driving tech from both the national and regional authorities continues to be extremely robust. That's the takeaway from our talks with local regulators, legislators, and officials. Their stance is still highly favorable. What's happening now amounts to a near-term recalibration. Essentially, they've paused granting fresh permits for additional autonomous vehicles. However, WeRide's self-driving cars are still out there running right now, and as I told you before, our bookings keep growing. When this technology shifts from small test programs into broader commercial deployment, it makes sense that regulators would zero in on safety. Honestly, we see that as a very responsible stance — it's absolutely the responsible thing to do, and we're completely behind it. Looking at it from where we sit, this ends up helping the industry over the long run. A firm that can point to strong safety performance ought to be rewarded, while one with a poor safety track record ought to face consequences. Think about how airlines work — that's exactly the same idea. The outcome doesn't change, correct? Airlines that maintain strong safety records keep flying, while those lacking one get pushed out of the market entirely. We're also suggesting that regulators ought to give preference to businesses with demonstrated technology, hands-on operating history, and excellent safety credentials. Over time, we anticipate that rules will grow increasingly tailored around things like safety results, operational history, and technical expertise. As for WeRide, our confidence holds steady given the substantial on-the-ground operating experience we've built up, in China as well as abroad. Here are just a couple of figures: we conduct testing and operations in more than 40 cities spanning 12 nations — has anyone else reached that milestone? Overall, we view this as something the industry has to do, and we believe it will ultimately lead to a more durable competitive environment. I don't want a business with a bad safety history operating in the market. That puts the public at risk, and it puts our community at risk. So we have a lot of confidence that safety — a company with a strong safety track record, the way WeRide has a higher motor, and WeRide is going to continue as a top participant, an early mover, in maintaining our strong safety history and maintaining our reliable operations. We make every effort we can to sustain this strong record. Let's move to the second one. My earlier response ran long, so I apologize for that. On this next question, I'd like to set the stage by noting it concerns a longstanding argument — the one pitting LiDAR against HD maps, and Waymo's strategy against Tesla's. My take is straightforward. I lived in Missouri for a good while — over a decade, twelve years or so. People have long referred to Missouri as the Show Me State. So for me, debating methodology isn't the point. Results are what I care about. When it comes to L4-level robotaxis, I believe it's clear to everyone at this point that Waymo has put thousands of robotaxis on the road and run them safely across numerous U.S. cities. Many different parties like to take credit, but my main point is this: when it comes to L4 robotaxis — vehicles that drive themselves with no one at the wheel — safety matters enormously. Should a high-definition map be usable as one source of information, there's no reason to avoid it; our company committed long ago to a multisensor redundancy strategy. Our conviction is that pairing cameras with LiDAR yields a powerful, dependable self-driving platform, and that stance is largely consistent with what the top robotaxi operators worldwide, Waymo among them, have chosen, correct? Meanwhile, we keep a close eye on how vision-centric and related methods are advancing, Tesla's latest moves included. I should say that FSD 14.3 represents a really solid step forward, and Tesla deserves credit for what they've accomplished. That said, it still isn't a robotaxi, and it isn't driverless. But — we chose a comparable path, which I brought up during the opening remarks, correct? Our ADAS offering, built on a single-stage end-to-end world model, has taken 4 straight titles in China — essentially, we've outperformed every other ADAS < > alternative out there. As far as I can tell, the one rival we've never gone up against directly is Tesla's FSD 14.3. And we're eager, as I said earlier, to line up side by side, head-to-head, and deliver the strongest possible experience for our customers. Let me stress one thing: we fully recognize how much leverage you get from camera-only approaches built on foundation and world models. And that's an area where we perform quite well. That's exactly why, in my view, I'm currently the CEO best positioned to take on this question. We've also given real thought to the extremely intricate traffic conditions you encounter in European cities like Madrid. We've already invested substantial effort mapping out every possible edge case in Madrid, because launching in a dense urban environment — a lively European metro area such as Madrid — demands that kind of preparation. Historically, somewhere like Zurich or Madrid would require a lot of localized work and testing of our assets. But with the GENESIS model, we feel good about actually bringing together the HD Map approach and the < > variant; that adaptability to different architectures is something WeRide does well; and we're — we can pull off a seamless combination of this HDMI LiDAR-plus-camera offering and the < > offering. That said, broadly speaking, what I really want to emphasize is that for a safe and reliable robotaxi rollout right now, we truly have to lean on HD Map. But if we take it one stage at a time, we could use the < > method for building emergency worlds and for updating the map, and we'd be able to keep the map current ourselves. Even so, generally speaking, what we need to do is take the best parts of the camera-based approach — the < > approach that relies on a world model, the way Tesla backs it — and bring those into our own core system. That's the strategy we're following, and as I see it, right now WeRide is the only company anywhere that can pull this off, thanks to our ADAS programs — our ADAS programs — and thanks to the operational experience we've gained from alpha-level deployment. All right. That's everything I wanted to say.

Analyst analyst
#8

Alright, so I have a couple of questions for you folks, okay? First, we keep noticing that a growing number of Level 2 portfolio companies are putting out announcements about their intentions to move into the Level 4 market. So how do you see that? And beyond that, could you share additional specifics regarding compatibility of the multi-chip platform for Level 2? From what I gather, this transition — so in what way do you pull off this chipset versus < >?

Xu Han executive
#9

Okay, thanks for those two questions. In my view, they lean more toward the technical side. So I will address both of them, alright? The first one concerns something like a general picture of the so-called L2 company, or ADAS company, even though those firms themselves would rather not be labeled an L2 company. They intend to step into our core market — what is our perspective? From where we sit, at present, I would still stress that to the best of my knowledge there is just a single company that does L4-level robotaxi while simultaneously running an L2++ system. I hold certain criteria for any company that brands itself a robotaxi company or < > company. You must possess a robotaxi fleet of 100 driverless robotaxis, available to the public, in operation for over half a year, and free of any significant or serious accidents. Excuse me, may I remind the listeners, if you aren't speaking, please mute your mic, since I picked up some background noise; so kindly mute your noise when you aren't speaking, okay; so I — so for a company at L4 level, you need a fleet of 100 vehicles open to the public, running for half a year. Only then may it call itself a robotaxi company; as for an ADAS company, you must get your system onto a mass production car. Earlier, I would say, 3 years back or 4 years back, WeRide, despite having certain ADAS solutions in our lab, still couldn't call itself an ADAS company, since we hadn't put our system onto a mass production car, yet today we can, because we have deployed our ADAS < > across more than — close to 30 types of vehicles, and several of those are selling by tens of thousands each month. alright? So now we understand the distinction. Having said all that, I believe a substantial gap remains between an advanced driver assistance system, meaning an ADAS system, and a genuine L4 robotaxi driverless system. The difficulty essentially comes down to robustness, operational capability and scalability. So here are a few numbers I would like to share with you, okay? So verbally, most data systems nowadays, based on our testing, mile per critical intervention. If you convert into kilometers, even though the term is MPCI, mile per critical intervention, many competitors in China, their target is to hit 1,000 kilometers of MPCI. Yet I want to give you a figure that for L4-level autonomous driving, you can look at numbers from WeRide and from Waymo. And the managed per critical intervention sits above the 1 million kilometers mark, which means we're looking at a 3 magnitude gap. So even though certain companies out there say they can increase their MPCI by a factor of 10 annually, they're still 3 years out from that. But I've got to tell you, multiplying your MPCI by 10x each year is an incredibly tough job — practically a mission impossible — because if any company could pull that off, then WeRide, Waymo, all these strong players who began autonomous driving back in 2017, think about where their MPCI would be after 7 years. So what I'm really getting at is you genuinely have to run for half a year before you understand what makes L4 so hard. It's wonderful to set an ambitious goal, but you always have to genuinely pursue those ambitious goals via a concrete path, and that concrete path matters most; based on what we've been through, it takes many years to truly get your fleet ready and dependable for driving and for operation. So my golden test rule is this: build a driverless fleet of at least 100 cars, bring it to market, run it for half a year, and watch what unfolds. And I don't want to witness any tragic names come up. You'll find some companies actually exit the market over certain accidents. You're all familiar with their names. So I'd rather not evaluate them any further. Okay. The second question concerns our multi-chip platform, and that's genuinely something I'm happy to address. I believe the core source — the secret sauce — is our GENESIS model. Through our GENESIS model, we create an AI model capable of trimming to various chipsets of varying complexity. And up to now, we've already launched our system built on NVIDIA's dual < > system. Plus, we launched our system, GAC N60, built on Qualcomm 8650, and we're about to launch a system built on < > on engines of chipset, okay? So fundamentally, we're confident that with our present GENESIS based approach, we establish a word model, and we can genuinely handle a range of different platforms, which hands us a significant competitive edge. And once more, as far as I know, we haven't come across any other company able to handle such a broad spectrum of chipsets simultaneously, making them work for ADAS solutions and for global taxi solutions too. Okay. That covers my response to both questions. Thank you.

Kai Xiao analyst
#10

I've got another question, this one about the multichip platform strategy that you brought up. Could you tell us what makes that strategy matter so much for your L2++ ADAS business?

Xu Han executive
#11

All right, so the basic answer is that when it comes to hardware preferences, OEMs aren't all the same — some of them are after chips with a really strong cost-performance ratio, while others are looking for a platform with very high tops computational power. Their supply chain needs differ too. What a flexible multichip architecture gives us, in essence, is the ability to serve a wide variety of vehicle platforms without having to redo the system design each time. And — though I should let the model speak for itself here — GENESIS is truly central to this, since it lets us handle a whole spectrum of onboard AI models, which covers varying top requirements such as 8650 < >. That means we need the capacity to adjust. Even more significant is the fact that flexibility across chip vendors makes mass production quicker and cost optimization stronger, since cheaper platforms come with differing costs and supply chain benefits; so that's one thing — when an ADAS company or an autonomous driving software company can offer a spectrum of solutions built on differing chipsets, car OEMs are inclined to collaborate with you more closely. We also think this kind of flexibility helps us lock in additional production and more vehicle types. Beyond that, we're eager to roll all of these systems out for the broad market. There's one thing I'd like to direct all the investors' attention to: during this year, exports of Chinese automobiles — Chinese mobile products — to the rest of the world are in fact expanding, yet the majority of them lack a very good ADAS system. And we have in fact already locked in more than < > vehicle models to supply ADAS systems for. So perhaps next year, you'll see some overseas models, and maybe you can drive a car equipped with the WeRide ADAS system in Mid East Asia, Southeast Asia, in certain countries, and even in Europe — please stay tuned. Thank you.

Xinyu Fang analyst
#12

My question concerns how you weigh and rank the strategic importance of China versus your business abroad. As everyone knows — and Tony brought this up — there's been reporting about stricter scrutiny of autonomous driving permits inside China, while WeRide keeps steadily moving forward in overseas markets with improved UE. Could management share its current thinking on how domestic and international operations are balanced and which takes priority?

Xuan Li executive
#13

All right, Xinyu, I'll handle that. In our view, China and the markets we operate in internationally both carry strategic weight for WeRide; in — Naturally, over the near to medium term, a number of international markets present a more obvious and quicker route toward robotaxi economics that are commercially appealing, given the pricing partnership that is favorable and < > such as supportive regulation, and proving sustainable profitability at an early stage matters greatly for our industry, since healthy cash flow makes self-sustained growth possible. And within local markets, that genuinely matters a lot. Getting to this point — what we call the < > kind of self-sustained growth — is very important. So we — we began establishing a substantial international footprint back in 2021, earlier than the majority of our peers; and since that time we've accumulated expertise in global deployment, regulation, localization, home location and fleet operations that is extremely difficult to copy. At the same time, China stays a crucial long-term market given the scale of its ecosystem infrastructure, and after all, it's where we're from. We keep a very strong preference there as well; so on the whole, we observe a potent flywheel effect as they scale into additional cities and additional countries, and we gain more data plus more validation to enhance performance and also to strengthen regulatory trust. So it speeds up the permits, audit expansion and commercialization. So yes, both matter a great deal to us. So naturally, we're delighted to say we foresee enormous potential for international revenue to rise quickly this year, and we remain very much on track to hit our full-year revenue target this year. Yes. That's everything from my side.

Tianyu Lu analyst
#14

Two things I'd like to ask. First, what percentage of Uber does WeRide currently have as a shareholder, and how should investors view the ties between WeRide and Uber? Second, would you walk us through your go-to-market approach in the various markets?

Xuan Li executive
#15

Broadly speaking, Uber holds above 5%; Uber functions as a strategic investor and a crucial partner for WeRide. According to their most recent public filing, their stake in WeRide exceeds 5%, which we see as a robust vote of confidence in WeRide's technology along with our dependable rollout and commercialization plan; so on the operational side, we're already putting Uber to work in a greater number of cities beyond the U.S. than any other AV company, which speaks to how deep our relationships run and our capacity to deliver at scale. So we anticipate that we'll grow — there seems to be an echo in the room, could somebody — can you mute? Thanks. Yes, we anticipate growing to — the echo is still there. Yes. We'll launch in additional cities alongside Uber during this year; and yes, so we continue to be a dependable ally as they scale up, as they build out their Global < > strategy. Naturally, meanwhile, our go-to-market strategy is multi-pronged. We team up with whichever partner suits each of us best in every market. And within China, we provided our own app and to establish < > recognition and operational strengths. You can hail our robotaxi through WeRide Go and you can also access it via the Amap and then the Tencent mini program as well. In Southeast Asia, we teamed up with Grab and in Europe, we work alongside local platforms, operators and PTOs, such as SBB in Zurich, TPG in Geneva and < > in Slovakia, and so on. So all of this will speed up our deployment and the permitting process. Yes. So on the whole, we see robust alignment with a platform such as Uber rolling out our robotaxi worldwide; and naturally, we keep taking a flexible approach in partnering with multiple platforms to make sure we execute very effectively across different market. And the second one, the other one is the asset — our asset-light business model for robotaxi Yes. I think the time < > short, I'll try to be quick. So beyond China, we already roll out an asset-light business model in every market. It's a mature and proven structure for WeRide; so in like — I'll just give you an example; in certain cases, robotaxis get purchased by the < > platform or like local customers; in certain other cases, they are owned by third-party fleet owners. We have already executed in both models successfully. What the asset-led approach means is that WeRide will concentrate on delivering the tag and the operation, while using local capital to grow more efficiently. And at present in China, what matters most is to keep enhancing the unit economics and, naturally, operational efficiency with a wider ODD, so that the revenue share grows sizable enough to appeal to a third party as an owner who would take part. So in the long run, we also expect China to head the same way; as utilization keeps getting stronger, the model ought to shift on its own toward a more asset-light structure.

Analyst analyst
#16

I'll keep it to just one question, then. What does the path look like for bringing down robotaxi vehicle costs?

Xuan Li executive
#17

All right. I'll handle that one; lowering costs, naturally, is something that stays among our main priorities, and we keep witnessing substantial headway, with 3 primary factors at play. The first involves hardware plus system integration. Thanks to our newest platform and the GXR, the vehicle we built for this specific purpose, we're heading toward a pre-integrated, highly standardized approach; and annually, BOM costs keep falling; the second factor concerns scale along with supply chain optimization. At present, we're shifting to deployments in the thousands of units. As is evident, we can achieve superior cost efficiency throughout various components and manufacturing. And the final factor relates to operational efficiency. We already observe a robust gain in efficiency as the fleet grows and utilization rises. To illustrate, during the previous quarter, we stated publicly that our ratio for remote safety officers had reached 1:40, up from 1:10, 1:20 before, and now it stands at 1:40. Abroad, the same pattern is unfolding, though today it isn't yet 1:40, we're headed in that identical direction. Thus, we consider each of these to be quite significant, since remote operational efficiency carries a very substantial effect on per vehicle TCO and on overall unit economics. On top of that, upon entering certain new markets, upfront items such as home litigation and localization costs may appear, which at times briefly push up per vehicle cost during the early stage. Yet as deployment scale grows within every market, such costs get amortized and their total effect turns out to be quite limited and manageable. So indeed, that covers cost reduction as a blend of hardware plus operational efficiency.

Operator operator
#18

We need to wrap things up because we're running short on time. I appreciate everyone joining us for this conference today; that brings the program to a close. You're free to disconnect at this point.

Read the full transcript via the API

You're viewing the first half of this call. Get the complete WeRide Inc. transcript - plus 132,000+ transcripts from 7,000+ companies, speaker segments and full-text search - through the EarningsAPI REST API or hosted MCP server.

Get an API key View API docs →

For developers and AI pipelines

Programmatic access to WeRide Inc. earnings transcripts and 132,000+ others is available through the EarningsAPI REST API and the hosted MCP server. Quarterly plans from $162 - full transcripts, speaker segments, full-text search, and the /api/v1/transcripts/recent polling endpoint for ETL pipelines.