Tesla participated in Hot Chips 34 and shared a ton of mind-boggling information about their Dojo supercomputer and chip architecture.
The details shared about Dojo in Hot Chips 34 by Tesla’s Emil Talpes, who worked at AMD for about 17 years for Opteron processors, are only about the hardware and capabilities of the tiles and Dojo as a whole. The performance will be discussed at Tesla’s AI Day II on September 30th.
The goal with Dojo is, according to Musk, “to be really good at video training. We have probably the fourth or approaching the third most powerful computing center in the world for AI training. Our first goal with Dojo is to make it competitive and be more effective and neural net training than a whole bunch of GPUs.”
Since Tesla needs a lot of computing power to process the video data from the vehicles in its fleet, it has built a proprietary system-on-wafer solution. According to ServeTheHome, “Each D1 die is integrated onto a tile with 25 dies at 15kW. Beyond the 25 D1 dies, there are also 40 smaller I/O dies.”
All of the power and cooling is integrated directly on the Training Tile, which is capable of 10 TB/s on-tile bisection bandwidth and 36 TB/s off-tile aggregate bandwidth. This architecture allows for the tiles to be scaled with 9TB/s links between them. They can also be plugged in and do not require their own server.
“The defining goal of our application is scalability,” Talpes said at the end of the presentation. “We have de-emphasized several mechanisms that you find in typical CPUs, like coherency, virtual memory, and global lookup directories just because these mechanisms do not scale very well when we scale up to a very large system. Instead, we have relied on a very fast and very distributed SRAM storage throughout the mesh. And this is backed by an order of magnitude higher speed of interconnect than what you find in a typical distributed system.”
The inside look into what Tesla’s building behind the scenes continues to prove how and why Tesla is at the forefront in artificial intelligence and neural net training. It gives the Tesla community an added sense of comfort knowing that Tesla will always have scalability and innovative technology at the forefront of everything the automotive company does.
Watch Anastasi In Tech’s Recap of Tesla’s Hot Chip 34 Presentation
Subscribe
Subscribe to our newsletter to stay up to date on the latest Tesla news, upcoming features and software updates.
Tesla has unveiled its 6-seat Model Y variant in China, known as the Model Y L. This new variant of one of the world’s best-selling vehicles comes with a longer wheelbase, adjusted C-pillar design, and most importantly, a six-seat interior layout.
The vehicle’s specifications have been officially listed in a filing with China’s Ministry of Industry and Information Technology (MIIT), confirming a launch for this fall.
Not a Tesla App
The addition of a longer wheelbase and a more spacious third row is a fantastic addition for the Model Y’s family utility, and positions this variant as sort of a mini Model X, but let’s compare the sizes to really know how this new Model Y compares to a Model X.
Meet the Model Y L
The defining feature of the new Model Y L is its six-seat configuration. This layout has previously been exclusive to the larger and more expensive Model X. While Tesla has offered the Model Y in a 7-seat configuration before, the third row was much too small to be utilized by anyone but small children.
Comparing Model Y L to the Model X
@xiaoteshushu on X
Let’s compare this upcoming Model Y L to the regular Model Y and the Model X.
Vehicle/Dimension
Wheelbase
Overall Length
Model Y
2,890mm / 113.8 in
4,797mm / 188.9 in
Model Y L
3,040mm / 119.7 in
4,976mm / 195.9 in
Model X
2,965mm / 116.7 in
5,060mm / 199.2 in
The new wheelbase of 3,040mm is a significant stretch from the standard wheelbase, and in fact, is longer than the Model X’s wheelbase of 2,965mm. However, the overall length of the vehicle is 84mm (~3 inches) shorter than the Model X. This means the vehicle sits neatly between the current Model Y and Model X, filling a much-needed gap.
While this Model Y L is slightly smaller than the Model X, it doesn’t necessarily mean that it’s smaller inside. The Model X features a much larger front end than the Model Y, accounting for several inches. When you line up the front wheel base of the Model X with this new Model Y, the vehicles are almost exactly the same length.
Tesla has designed this Model Y to be a bit more compact and efficient than the Model X, and likely much cheaper, while featuring the well-loved design of the new Model Y.
Other Specifications and Price
The MIIT filing also provided a detailed look at some additional specifications. The Model Y L is a dual-motor, AWD variant, so it will likely be more expensive than the current Model Y AWD that’s available in China today. Tesla charges an additional $6,500 USD when upgrading the Model X from a 5-seat configuration to a 6-seat layout, so we may see something similar here.
The extra length has been added behind the C-pillar, resulting in a longer rear profile for the Model Y L. To accompany this, Tesla has added an updated rear spoiler, similar to the one found on Performance variants, but not carbon fiber. There is also a new wheel design to complement the updated look, along with unique Model Y L badging and a new light gold paint option.
In classic Tesla fashion, no Tesla is slow - and the Y L has a 0-100 km/h (0-60mph) time of 5.9s, with a top speed of 217km/h. Alongside an 82.5 kWh LFP battery pack, the Model Y L boasts an impressive CLTC range of 688 km (427 mi).
Not a Tesla App
Launch & Availability
According to posts from Tesla China on Chinese social media, the new Model Y L is scheduled to launch in the fall of 2025. Its official listing in the MIIT database is essentially the final regulatory step required before sales can begin, which means the launch is really just around the corner. For now, it appears that Tesla intends to launch this vehicle only in China, as no other filings have been made in other regions. However, these could be revealed in the coming months.
The new Model Y L is a huge addition to Tesla’s lineup - one that addresses the Chinese preference for vehicles with longer wheelbases and additional passenger room in a compact SUV package. The question is - will this variant make its way to North America and Europe?
Solving real-world artificial intelligence - whether for autonomous driving, real-world robotics, or advanced reasoning - requires an almost unfathomable amount of computational power. To meet this challenge, Tesla has been developing its own custom AI training hardware while simultaneously purchasing hardware in the open market.
Now, the next-generation Dojo 2 chip has reportedly entered mass production with the world’s largest semiconductor manufacturer, TSMC. While many may consider this a side quest, expanding Tesla’s computing base will be necessary to achieve exascale supercomputing, which will be crucial for all of Tesla’s AI ambitions.
Elon Musk called Dojo 2 “a good computer,” and then followed up with a classic computer performance joke - Dojo 2 can indeed play Crysis at a billion frames per second.
While Tesla has effectively utilized powerful third-party GPUs to train its models to date, the Dojo supercomputer is a ground-up, application-specific solution designed for a single purpose. It will efficiently process massive amounts of video data for training neural networks. The Dojo 2 chip itself is the key that unlocks this potential.
Dojo 2 will train the vision-based neural nets that FSD relies on, allowing Tesla to process video from its massive global fleet of vehicles even faster. As Tesla continues to improve FSD, one of the biggest challenges has been the intake of video for handling difficult edge cases.
Hundreds of thousands of miles of training data may pass by before an edge case is identified and trained on, but it all needs to be analyzed, labeled, and processed, which is key for Dojo 2. Each new useful piece of training data will help Tesla proceed down the march of 9s, making FSD just that little bit better every time.
This process requires massive amounts of compute and training time - but it is an absolute necessity to improve FSD. Of course, this goes beyond just FSD in vehicles. Tesla’s humanoid robot, Optimus, also runs on FSD to navigate and interact with the physical world.
While it may be a custom version of FSD, it remains FSD at its core, which means the same neural nets that analyze the environment and build a 3D map of the world for your car perform the same work for Optimus.
Not a Tesla App
Custom Approach to AI Hardware
Dojo 2’s power doesn’t just come from raw compute; it comes from a series of architectural choices that make it excel at training FSD and differentiate it from general-purpose hardware, or even other AI-specific hardware.
To this end, Tesla is using TSMC’s new Integrated Fan-Out with Silicon-on-Wafer (InFO-SoW) packaging technology. For massive AI workloads, heat and the speed at which data moves between chips are often the biggest bottlenecks.
This new packaging technique allows for high-bandwidth connections directly between processing dies, which lowers latency and dramatically improves heat dissipation, all key to building massive and dense compute clusters.
Unlike general-purpose chips, Dojo 2 is designed with a custom instruction set, specifically built to train FSD. The cores are specifically made to accelerate the exact mathematical operations, like matrix multiples and systolic arrays, which form the backbone of Tesla’s vision-based neural networks.
By building its own hardware, Tesla can then integrate its own software and compilers directly with the silicon, optimizing for specific workloads and avoiding the performance penalties that can result from using third-party software, such as Nvidia’s CUDA.
The start of Dojo 2 may seem like a side quest for some, but it’s actually a key step for Tesla’s AI technologies that give them an advantage over the competition using off-the-shelf hardware. They’ll need to continue investing in custom hardware to improve FSD at a reasonable pace, rather than the current glacial pace we’ve seen over the last few months.
![Tesla Software Update 2025.26: Grok, Light Sync, Audio Presets and More [VIDEO]](https://www.notateslaapp.com/img/containers/article_images/tesla-update/light-sync-update.jpeg/e84ab2f1fe12f493a75927db105a9586/light-sync-update.jpg)
