Tesla is planning various improvements to FSD Beta that could drastically increase its performance
As we explore the future possibilities of Tesla's Full Self-Driving (FSD) Beta, we enter a realm of speculation and excitement fueled by hints and ideas shared by CEO Elon Musk. From integrating neural nets in vehicle control to the unique concept of "reverse creep" and independent navigation, Tesla's FSD Beta could be on the verge of a massive transformation.
Future Expansion of Neural Nets in FSD Beta
One of the most speculated advancements in Tesla's FSD Beta is the comprehensive application of neural networks for vehicle navigation and control. This move beyond their current use for vision processing could revolutionize autonomous vehicle technology, empowering FSD Beta vehicles to make informed decisions that more closely mimic human cognitive processes.
Reverse Creep: A New Dimension to Autonomous Safety
Imminent updates to FSD Beta may also include the intriguing capability of "reverse creep." This feature would enable the vehicle to move backwards when it senses potential danger, mimicking the cautious maneuvers of a human driver in a challenging driving situation. By extending its response beyond just creeping forward for improved visibility, FSD Beta could improve autonomous safety.
Towards Map-Free Navigation
Another potential leap for Tesla's autonomous driving tech is the ability of FSD Beta to navigate without relying on map data. This would allow Tesla vehicles to tackle even the most remote or poorly mapped routes, liberating them from the confines of pre-existing cartographic information. All it might need is a set GPS point or pinned location to embark on its journey.
Dead Reckoning Navigation: Pioneering GPS-Free Movement
Speculation is rife about the development of 'dead reckoning' navigation. This advanced feature could enable FSD-equipped vehicles to navigate based on inertial measurements, wheel movement, and vision, even when GPS data is unavailable. As such, Tesla vehicles could find their way through environments like underground parking garages, relying on their last known GPS location and determining subsequent positions using a combination of compass data, wheel movement, and speed.
According to Elon Musk, these are enhancements that Tesla is working on to improve FSD. They've been expected in previous versions of FSD Beta, but have likely been delayed or are still in development.
These features help showcase the dynamic trajectory that Tesla's Full Self-Driving Beta is possibly heading towards. As we look forward to these and many more breakthroughs, the future of autonomous driving promises to be nothing short of revolutionary.
Tesla has been quietly building something pretty unique on the streets of Hollywood—and we’ve been keeping a close eye on it. That’s right: the long-anticipated Tesla Hollywood Drive-In, Diner, and Supercharger is shaping up to be one of the most creative approaches to EV charging yet.
But this isn’t your average Supercharger site. It features a retro-futuristic, 24-hour diner and an outdoor movie theater, blending the charm of the 1980s with the tech-forward vibe of the 2020s.
Tesla hacker Greentheonly managed to spot some new evidence of changes within Tesla’s 2025.8.6 software build, which includes a new modal for “Charger Diner Controls”. While there isn’t anything to activate this new modal just yet, it is an indication of how Tesla intends to integrate the experience right into each and every Tesla.
The text in the modal reads: “Plug in, place your order, recharge” - which means that once you arrive and are plugged in, you’ll be able to place your order - and likely pay through Tesla’s payment system. It seems like this will all be done through either the Tesla app or the vehicle’s tocuh screen. No need to fumble with a separate app or pull out your wallet.
This is definitely a level of integration you just can’t have anywhere else - and just like any other Supercharger site, the Tesla Diner will appear as a suggested site for charging when you’re navigating or looking around the Hollywood area.
This is our best look yet at the two 45-foot LED screens at @Tesla's upcoming Hollywood diner and Supercharger.
We can also see the second Supercharger lot on the left with solar canopies. This will be one of the largest Supercharging spots in North America. pic.twitter.com/aWUe67M5vw
Those two absolutely gigantic 45-foot LED screens have now been turned on for the first time, meaning that Tesla is likely close to opening the Supercharger site. The screens were spotted displaying some test patterns, as well as some Tesla-themed content. It really brings the construction site to life - and it's the clearest sign that the Diner is well on its way to completion.
We’re hoping that this first Tesla Diner sees plenty of success - and that Tesla begins opening more of these types of facilities throughout North America, and eventually globally. Given Tesla’s penchant for unique Superchargers, we imagine this could truly be the case one day.
As Tesla continues to push the capabilities of its hardware as it prepares for autonomy with FSD Unsupervised, there has been a distinct divergence in features between HW4-equipped vehicles on FSD V13, and HW3-equipped vehicles on FSD V12.6.
There are definitely reasons for this, and Tesla hacker Greentheonly, over on X, went into great detail on some of the reasons why this divergence has continued and is likely to continue.
If you’re interested in some background, we recommend the following articles. Although they’re not necessary, you may enjoy reading them before diving right in if you enjoy this topic.
Back in the summer of 2024, we reported that Tesla’s HW4 had yet to reach its full potential, and that HW3 still had some compute to work with. However, FSD has undergone a rapid evolution - there are now fundamental differences driven by the hardware capabilities and sheer scale of the latest builds of FSD.
FSD V13’s Gargantuan Neural Networks
FSD V13 is a substantial leap for Tesla. It features substantially larger neural networks - the core brain that processes all the environmental data and makes decisions. Those networks are far larger in V13 than in V12, which results in a much smoother and human-like expereience.
Tesla’s FSD computers contain two nodes. According to Green’s analysis of FSD V13, Node B, which handles the actual end-to-end driving logic, tripled in size from 2.3 GB in V12 to a staggering 7.5 GB in V13. That’s a substantial increase, and one of the main reasons why HW3 can’t handle FSD V13. While Interestingly, Node A on HW4 remains similar in size to FSD V12, using up about 2.3 GB.
However, this massive size is also pushing the limits of HW4. While this isn’t necessarily a big problem immediately - at the end of the day, there is a limited amount of compute and ram available and model sizes will likely grow in the future. In fact, Tesla is planning to scale model sizes in the next FSD version and has already said they’ll need to optimize them, as context size is limited by on-board memory.
HW3 Limitations
Since HW3 can’t run the latest FSD models, where does this leave millions of Tesla vehicles equipped with the now-legacy FSD hardware? Are they hitting a dead end? Not exactly - Tesla has promised to replace HW3 computers with new units during the Q4 2024 Earnings Call, but only for those who have purchased FSD and only if Tesla is no longer able to update HW3 vehicles.
There’s no doubt Tesla hopes to achieve autonomy and then work backwards to make it compatible with HW3, but that’s unlikely to happen with the limited ram available on HW3.
Tesla’s AI team has continued to say that they’re still working on updating HW3 vehicles - but that they will be taking a backburner to updates for HW4 until the team can optimize the model efficiently enough to run on the smaller footprint. That means V12.6 isn’t just V13 “lite” - it is a distinctly different software branch tailored specifically for HW3’s capabilities.
Green's analysis showed that HW3 under V12.6 runs NNs totaling 1.2 GB on Node A and 3.1 GB on Node B, which is significantly smaller than the footprint of V13. Back in November, we discussed how Tesla is running a smaller and optimized FSD model on HW3, with the specific goal of achieving similar performance to V13. FSD V12.6 appears to be the current implementation of that smaller model strategy.
Green also found that of the total neural nets used for FSD, 135 are shared between the current V12.6 and FSD V13.2 releases, which is the result of Tesla’s AI team working to optimize and refine the V13 model for HW3. So some neural nets are shared between the two sets of hardware, while others are specific to HW3 or HW4.
AI4 in the Future
The fact that V13 is already stretching HW4’s limits makes us wonder about Tesla’s imminent release of AI5. Last year, Tesla mentioned that AI5 would be 18 months away - which is now about 8 months away, and potentially even sooner to help solve the autonomy problem.
At the same time, Tesla is looking to launch Unsupervised FSD in June in Austin, Texas, with presumably HW4, which would indicate that they’re confident in its capabilities.
Unless Tesla launches the Robotaxi network with AI5, we don’t believe AI5 will be a “must have” item. Undoubtedly, it will result in similar improvements to HW4 compared to HW3. However, the software needed to take advantage of the newer hardware could still be years away.
All eyes will be on Tesla’s Robotaxi network launch and the hardware required for it. For now, HW4 owners are in good hands with future updates are on the horizon, and we’re certainly excited for what’s next.
![Tesla’s Hollywood Diner: In-Car Controls & Theater Screens Turned On [VIDEO]](https://www.notateslaapp.com/img/containers/article_images/2025/tesla-diner.webp/3aaa2ea0edf2eaa0fabe1217530f3476/tesla-diner.jpg)
![Tesla’s Missing Voice: Why a PR Team Matters More Than Ever [Opinion]](https://www.notateslaapp.com/img/containers/article_images/multiple-models/group_93.jpg/4e1056961f0480c7b9eff43dd2ec288e/group_93.jpg)
