Xpeng VLA 2.0 matches Tesla FSD in Beijing test drives, signaling shift in autonomous driving race

Xpeng VLA 2.0 test drive: Tesla is not alone with 'Full Self-Driving' anymore

I test-drove Xpeng's new VLA 2.0 autonomous driving system in Beijing last week, and after 40 minutes of navigating one of the most aggressive driving environments in the world, I didn't have to intervene once. This is a major development. Tesla has been the clear leader in advanced driver-assist systems in consumer cars, but Xpeng's latest update proves that lead is shrinking fast -- and in some ways, it's already gone. Xpeng rolled out its VLA 2.0 (Vision-Language-Action) system starting in March 2026, delivering it via over-the-air updates to its latest vehicles, including the P7, G7, and X9 in their "Ultra" configurations. VLA 2.0 represents a fundamental architectural shift from Xpeng's previous NGP system. Instead of relying on separate perception, planning, and control modules, VLA 2.0 uses an end-to-end vision-to-action model that translates what the cameras see directly into driving decisions -- eliminating the intermediate translation layers that can introduce latency and errors. The system is powered by Xpeng's proprietary Turing AI chip, delivering up to 2,250 TOPS of computing power on production vehicles. Xpeng trained the model on 100 million clips from "extreme driving scenarios," and it shows. The company says driving efficiency improved 23% over the previous generation, with 99% fewer hard braking events. The architecture is similar to what Tesla has done with its end-to-end neural network approach for FSD, but Xpeng has been iterating rapidly. Volkswagen was so impressed that it signed on as the first external customer for VLA 2.0, deploying it in its new electric SUV for the Chinese market. I tested the previous generation of Xpeng's system in Guangzhou last year and was impressed, but VLA 2.0 in Beijing is a different experience entirely. Beijing traffic is ruthless. Drivers cut in aggressively, lane discipline is more of a suggestion than a rule, and yielding to merging traffic is seen as weakness. If your self-driving system hesitates, you'll never make a lane change. VLA 2.0 handled it. During my 40-minute test drive, I didn't have to take over once. The system navigated complex intersections, managed aggressive merging situations, and kept up with the flow of traffic without being overly cautious or dangerously assertive. There was one moment that stood out. The vehicle needed to merge into a tight gap in heavy traffic. In most driver-assist systems I've tested, this is where the car would either hesitate until the opportunity passed or require a human takeover. VLA 2.0 committed to the gap, asserting itself into the lane the way an experienced Beijing driver would -- firmly but smoothly. It was impressive. I won't argue that it didn't make me nervous and I was fairly close to pressing the brakes, but it worked out. You can tell this system was trained for Chinese road conditions. It doesn't drive like a cautious American suburbs algorithm dropped into Beijing chaos. It drives like it belongs there. In my short 40-minute drive, VLA 2.0 felt like driving my Tesla on FSD v14. That's not a comparison I make lightly. The comparison to Tesla is inevitable, and Xpeng CEO He Xiaopeng is leaning into it. He traveled to Silicon Valley late last year to test Tesla's FSD v14.2 himself, spending about five hours driving in San Francisco. He called it "near-Level 4" performance -- high praise from a competitor. But He Xiaopeng didn't just compliment Tesla. He set a target: Xpeng's VLA system must match FSD v14.2's performance in China by August 30, 2026. He even made a bet with Xpeng's head of autonomous driving, Liu Xianming -- if they miss the deadline, Liu has to run naked across the Golden Gate Bridge. Obviously tongue-in-cheek, as I'm pretty sure this is not legal. The tension between the two companies runs deeper. Tesla's latest FSD software is not approved in China yet. Chinese Tesla owners who have access to advanced driver-assist features are running v13, while North American owners are on v14. Tesla had AI training infrastructure in China as of early 2026, but full FSD approval keeps getting delayed -- China shut down Elon Musk's claim that it would be approved in February. Based on what I experienced, Xpeng is already very close to matching FSD v14. The August deadline feels achievable. This test drive confirmed something we've been tracking for a while: Tesla is no longer the only company shipping extremely advanced Level 2 driver-assist systems that can navigate city streets in consumer cars. VLA 2.0 doesn't just work in some sanitized demo environment. It works in Beijing -- arguably the toughest urban driving environment you can throw at a self-driving system. That matters. There's also the pricing question. Xpeng includes VLA 2.0 in its vehicles. Tesla charges $99 per month for its FSD subscription in the US, with the company signaling the price will increase significantly as capabilities improve. Tesla moved to subscription-only pricing earlier this year, eliminating the one-time purchase option. Xpeng has indicated it might charge for VLA at some point, but for now, it's included. Either way, it's increasingly clear that charging a significant monthly subscription for these features is a challenging business model when competitors are including comparable technology in the price of the car. And Xpeng is not alone in challenging Tesla here. BYD's "God's Eye" system comes standard on vehicles priced around $30,000. Huawei is investing $11 billion to $13 billion in autonomous driving software over the next five years. Xiaomi is pushing hard on smart driving in its SU7 lineup. The field is getting crowded, and all of these systems are available in China -- the world's largest EV market -- where Tesla still can't run its latest FSD software. Tesla's technology remains impressive, but the moat is getting shallower by the month.

After using the Xpeng P7 Ultra's hands-free system, I'm convinced it's way better than Tesla's FSD | Stuff

Xpeng's VLA 2.0 might only be available in China right now, but it's more than a match for Tesla's full self driving When Xpeng suggested I try a hands-free autonomous drive around the roads close to the Beijing Autoshow I was keen, but also rather nervous. I'd already witnessed so much chaos on those multi-laned highways that being behind the wheel with jetlag and little experience felt borderline nuts. A lack of lane discipline is par for the course when driving around Beijing. However, a definite upside was being given the Xpeng P7 Ultra to try. This super cool electric car comes from a company that proudly boasts it is a tech business that just so happens to build vehicles. The P7 Ultra looks awesome from the outside, but the really impressive stuff is indeed the tech. Xpeng unveiled its VLA 2.0 automated driving system earlier this year. It works using a combination of cameras around the car combined with bespoke software and lots of computing power. The result is a dynamic hands-free experience that was a revelation. I managed to complete an entire test drive in proper hands-free fashion, only having to touch the wheel when prompted to let the system know I was still there. A couple of days prior, I had to spend a few hours at a Chinese driving licence centre to get my paperwork cleared and pick up a laminated driving licence after filling in lots of forms, getting my photo taken and sitting through a loop tape of terrible crashes. The rest of the class looked on very seriously as I sat there thinking how much the tape reminded me of Alan Partridge's 'Crash, Bang, What a Video'. Nevertheless, this bumper collection of mangled wreckage and bloodied bodies did underline how serious I needed to take driving in China. As it turned out, it was the Xpeng P7 Ultra that would do all the hard work. Slipping behind the wheel in the car park of my hotel, the instructor next to me gave a brief overview of what was going to happen and zoomed out of the map to show an overview of the loop we'd complete. I had it in my head that I'd be needed to supply some input for the driving part. However, save for those occasional taps of the steering wheel, VLA 2.0 did all the heavy lifting. This was a truly impressive demonstration. The car slipped seamlessly out of the parking lot and into the flow of traffic. At first, it was hard to sit back, relax and enjoy the drive as other cars, a few trucks and lots of mopeds attempted to block the route ahead. The Xpeng P7 Ultra took it all in its stride and not once did I have to grab the wheel or stab the brakes. By the time we'd got about halfway around the route, I started to relax a little and marvel at just how good this system was. Occasionally, I'd wonder if the system was going to forget to indicate and carry out a manoeuvre but, nope, every nuance in the route plan, which had a great mix of different roads, was taken without fuss or bother. I loved the way VLA 2.0 worked so dynamically too, providing an extremely smooth experience that involved a subtle tweak of the throttle here and a soothing dab of the brakes there. Overall, it was seamless. We topped off the experience by pulling back into the hotel car park, hopping out and letting the Xpeng P7 Ultra reverse itself into a parking space. Of course, in a place where patience is limited and horns are plentiful, not everyone seemed to be as impressed by Xpeng's tech as I was. But it was hard to not stand there thinking just how much of a gamechanger VLA 2.0 could be. However, the world is full of different drivers and countries all have their own driving rules and regulations. Working out how to get an Xpeng P7 Ultra to behave just as impressively in, say, the UK or Europe is an altogether different can of worms. Somehow, though, it's hard to see how the Chinese 'can do' attitude will stop it happening even if the legislation continues to say no. Xpeng is already testing the VLA 2.0 system in Europe and it's exciting to think of the opportunities such a good system could create for driving in some of the world's most congested and stress-inducing capital cities. All we need to do is wait, but there's more to it than providing us with smart products, as Elon Musk realised when the formative roots of the now huge Tesla network were being devised. Tesla pulled off a masterstroke when it not only came up with an electric car range but supplemented it with a worldwide charging infrastructure. Up until now, that has been a hard act to follow, but Xpeng is already making headway by developing its own global charging network, starting in China but with plans to extend the strategy to dozens of other countries. "We're creating a worry free and hassle-free experience for our customers," says Brian Gu, Vice Chairman and President of Xpeng at the Beijing Auto Show a day later. "From the product perspective, though, I think our focus is beyond just electrification. I think for our customers to really enjoy what the technology can bring to their lives, we are increasingly looking at the intelligence. We wanted to build an intelligent mobility ecosystem like you've experienced." "We have developed a core of what we call physical AI," he explains. "Utilising our chips, our models, our dataset and using our own electronic architecture that really comes together to deliver a more intelligent solution. You're not just thinking about driving an EV, you're actually driving something that can help you, make the experience safer and more enjoyable and more relaxing." Considering how calm I felt after my drive in the Xpeng P7 Ultra, it was hard to disagree.