Tesla Releases FSD V13.2: Adds Ability to Reverse, Start FSD from Park, Autopark at Destination and Much More

December 1, 2024

By Karan Singh

DirtyTesla/YouTube

Last night Tesla finally launched FSD V13.2 with a bevy of new features for its early access testers with update 2024.39.10. While they barely missed Thanksgiving's floaty deadline, they still managed to deliver it in November, marking another big win for the Tesla AI team.

Early Access Only

FSD V13.2 started to roll out to early access testers - who generally get hands-on with the latest builds in advance of everyone else. They’re the equivalent of Tesla’s trusted testers who aren’t running internal builds - and they’re able to catch more scenarios outside of Tesla’s pretty extensive safety training suite.

If no major issues are spotted, Tesla will begin a slow rollout to more and more vehicles over the next few weeks. Assuming all goes well with this build, it could be in most customer’s hands by Christmas.

Of course, as a reminder, FSD V13 is still limited to vehicles equipped with AI4—and for now, anything but the Cybertruck. The Cybertruck is on its own FSD branch, without access to Actually Smart Summon and Speed Profiles, but with End to End on the Highway. The Cybertruck was recently upgraded to update 2024.39.5 (FSD V12.5.5.3).

FSD V13.2 Features

Let’s take a look at everything in FSD V13.2 - which is the build version going out now on Tesla software update 2024.39.10. While we previously got a short preview of what was expected with V13, we now can see everything included in FSD V13.2.

Start FSD from Park, Reverse & Park at Destination

Parked to Parked has been the goal for FSD for quite a while now. Elon Musk has been saying that it was going to be the key to demonstrating Tesla’s autonomy framework back with the release of V12.3.6 - when V12.5 was but a glimmer in the Tesla AI team’s eye.

Now, with V13, FSD has integrated three key functionalities.

Unpark: FSD can now be started while you’re still parked. Simply set your destination and tap and hold the new Start FSD button. The car will now shift out of park and into drive or reverse in order to get to its destination.

Reverse: FSD has finally gained the ability to shift. Not only can the vehicle go into reverse now, but it can seamlessly shift between Park, Drive and Reverse all by itself. It can be perform 3-point turns.

Park: When FSD reaches its destination, it will now park itself if it finds an open parking spot near the final location. Tesla says that further improvements are coming to this, and drivers will be able to pick between pulling over, parking in a parking spot, driveway or garage in the future.

If everything goes smoothly on a drive, users will no longer need to give the vehicle any input at all, from its original location to its final parking spot. No more user intervention other than supervision is needed, unless an intervention is needed.

I tried to get FSD 13.2 to park in my garage but it instead did a 3 point turn and tried to escape 😱 pic.twitter.com/EzEYcNvuuA
— Dirty Tesla (@DirtyTesLa) December 1, 2024

Full Resolution AI4 Video Input

Until now, FSD V12.5 and V12.6 have been using reduced image quality at reduced framerates to match the lower resolution and lower refresh rate provided by Hardware 3 cameras. For the first time, FSD will be using AI4’s (previously known as Hardware 4) cameras at higher resolution and 36 frames per second.

In short, that means better image quality for both training and in use and higher accuracy for things like signage and distance measurement.

Speed Profiles for All Roads

FSD V12.5.6.2 brought new and improved Speed Profiles to both city streets and highways, including the new Hurry Mode, which replaced Assertive Mode. However, on V12.5.6.2, there were a few limitations - roads needed a fairly high minimum speed limit of 50mph (80km/h) or higher. Now, that’s no more. City Streets has speed profiles for all speed limits now.

FSD 13 navigates parking lot then BACKS ITSELF INTO SUPERCHARGER STALL pic.twitter.com/f6vAeJ0mSe
— ΛI DRIVR (@AIDRIVR) December 1, 2024

Native AI4 Inputs and Neural Network Architecture

Similar to the video resolution and refresh rate, AI4 has a lot of new hardware features that help optimize how fast FSD’s AI model can run. We dug into how Tesla’s Universal Translator streamlines FSD for each platform - this is a case of having fewer constraints and more optimization versus Hardware 3.

5x Training Compute

Cortex, Tesla’s massive new supercomputer cluster at Giga Texas, is now online and crunching data at a truly insane rate. It's one of the fastest AI clusters in the world—and it's dedicated to FSD. Tesla has 5x the training compute crunching away to solve the March of 9’s now that FSD is close to being feature complete.

Faster Decision Making

Tesla refactored how it handles image-to-processing in FSD V13 - another huge set of changes to improve performance. In this release - a 2x faster photon-to-control latency, which is massive. In layman’s terms - that’s faster decision-making - it was already faster than a human, and now it's twice as fast as it was before.

Insane product push on the week of my birthday.

We refactored the entire system to drastically simplify the pipeline — direct photons to control — yet providing a lot more functionality under the same unify framework.

It is probably one of the biggest rewrite in years when we… https://t.co/M0pr2KzcHv
— Yun-Ta Tsai (@YunTaTsai1) November 30, 2024

Collection Data for Audio Input

One of the features Tesla lists in FSD V13.2 is the ability for the vehicle to collect and share audio snippets with Tesla. The vehicle will ask you whether you’re okay with sharing 10-second audio files with Tesla so that the vehicle can detect emergency vehicles by sound in the future.

Camera Visibility Detection

The vehicle will now prompt you at the end of a drive if visibility issues are detected. The new option is under Controls > Service > Camera Visibility. Tesla will also retain images from the cameras when the vehicle experiences visibility issues during a drive so that you can analyze them later.

As of FSD V12.5.6.2, your Tesla will warn you when it needs cleaning - and guide you to help clean the cameras too. This, along with less annoying notifications that FSD is degraded, are going to be fantastic changes for those who aren’t driving around in sunny weather.

Better Collision Avoidance

Due to all the changes to the AI model in V13, it also brought along with it changes to how the AI perceives and handles collision avoidance.

FSD has already earned a reputation for cleanly avoiding T-bone collisions in red light incidents, but it's going to get even better from here on out.

FSD 13 leaves parking lot (+ awkward interaction with other driver)

the smoothness is absolutely INSANE

it also saw the Model 3 backing up before I did, I was wondering why it wasn’t moving lol pic.twitter.com/MhKgVI2WX1
— ΛI DRIVR (@AIDRIVR) December 1, 2024

Vehicle to Fleet Communication

One of the features V12.5 was supposed to bring was fleet-based dynamic routing. If a route was closed, your Tesla would turn around and navigate through an alternative path - and also warn the rest of the fleet of the closure.

V13 lets AI4 vehicles do this, and it's another element of the Robotaxi network that Tesla needs to get off the ground to ensure that once they do begin to deploy their first fleets - they function well. So far, with new job postings for Robotaxi Engineers and talks with Palo Alto to launch a Robotaxi service, things are on track for both Unsupervised FSD and Robotaxi sometime in 2025.

Better Traffic Controller

Another big update is a redesigned traffic controller - which makes for smoother and more accurate tracking of other vehicles and objects around the vehicle. We dug into how the traffic controller processes information in this article here, where you can learn all about how Tesla’s signal processing works.

Here is highlight clip #2 from my FSD Supervised v13.2 Night First Impressions Drive. Enjoy the clips! @Tesla_AI pic.twitter.com/bLOjM0bVYi
— Chuck Cook (@chazman) December 1, 2024

Upcoming Improvements

Tesla has mentioned a lot of upcoming improvements panel for FSD V13 too, which includes bigger models, audio inputs, better navigation and routing, improvements to false braking, destination options, and better camera occlusion handling. That’s a pretty big list for V13, so we’ll keep an eye on all these upcoming features that are expected in a future release.

What About Hardware 3?

Tesla’s previous roadmap update didn’t mention HW3 getting FSD V13. Instead, those of us on Hardware 3 will need to keep waiting and looking for Tesla to optimize another FSD Model - until then, you’ll be on FSD V12.5.4.2, which is still a fairly capable build.

Tesla has mentioned that they could potentially upgrade HW3 computers - not cameras - if engineers aren’t able to get FSD Unsupervised working on HW3. While there isn’t a lot to share here yet, it certainly looks like HW3 owners will be receiving some sort of free hardware upgrade in the future, but it’s not clear yet when or what they will be.

Keep an eye out in the new year for updates on what’s coming next with HW3. We hope to see an optimized V13 build eventually make its way to HW3 sometime in the future - Tesla has been working pretty hard on this, so let’s give them some time.

Release Date

For everyone who’s been patiently waiting to see more of FSD V13 since the sneaky reveal at We, Robot, you’ll be waiting a bit longer. This build is currently going out to early access testers, who serve as a critical step in Tesla’s safety verification process.

Once Tesla is comfortable with the rate of disengagement, Tesla will evaluate their results, make any final changes, and then begin rolling it out in waves. Fingers crossed, wider waves for V13 will make their way to AI4 S3XY vehicles and the Cybertruck by Christmas.

Tesla Building Cortex 2.0 Supercomputer at Giga Texas to Power FSD

May 14, 2025

By Karan Singh

Not a Tesla App

FSD’s insatiable appetite for AI compute is taking shape once again at Giga Texas, where Tesla is building out Cortex 2.0 — its second large-scale GPU supercomputer cluster — on the facility’s north side.

This confirmation comes thanks to the detailed research and findings by Giga Texas drone pilot Joe Tegtmeyer, who shared his findings on X.

For months, there has been some speculation on what this new site would be - new storage, the Optimus production line, preparation for the Cybercab’s unboxed assembly process, and more. Now, thanks to permits filed by Tesla, we know that this new area is destined to become the second supercomputer destined for FSD training.

Cortex 2.0

There are some key details we can unpack from what we’ve learned about Cortex 2.0. The new facility will be on the north side of Giga Texas, opposite Cortex 1.0, which is on the south side. The permits explicitly tie the usage of this expansion to Cortex 2.0, namely through the fact that it is a data center.

This is actually a change from Tesla’s original plans - the northern section was intended to be used as a central campus support facility, with three smaller facilities and water storage tanks. Now, it has been redesignated as Cortex 2.0 and is one large structure.

The permits that have been filed have already been approved and cover the foundations, underground water mains, and building itself. The steel structure is actively being put together on the concrete foundations, and about 50% of the roof decking is already complete, bringing the exterior shell closer to completion.

Fueling FSD’s Brain

The deployment of Cortex 2.0 is the latest in Tesla’s massive and ongoing investment in the computational power required to train FSD. While this primarily covers FSD for cars, this will also eventually apply to Optimus, which is powered by the same AI4 computer and also runs a variant of FSD.

This isn’t Tesla’s first foray into large-scale AI infrastructure and won’t be their last. Alongside Dojo, their home-grown AI supercomputer, Tesla has been partnering closely with Nvidia to ensure they have the GPUs necessary to do all the hard work. Cortex 2.0 is the next iteration, and once it's fully online, we can expect that Tesla’s work on FSD will accelerate even further.

Especially since Elon is planning something truly outrageous once again:

Been thinking about the fastest way to bring a terawatt of compute online.

That is roughly equivalent to all electrical power produced in America today.
— gorklon rust (@elonmusk) May 12, 2025

You can check out Joe’s full video below:

Tesla FSD in Europe: Highway Approval Expected to Arrive in September

May 14, 2025

By Karan Singh

Not a Tesla App

With European Tesla owners eagerly awaiting any news on FSD in Europe, we’ve seen Tesla tease FSD, and also point out exactly what the barrier has been - regulatory approval. Now, following the latest meeting of UNECE, new regulatory amendments are set to unlock “System-Initiated Maneuvers” (SIM) on highways across participating European nations.

This development, highlighted by Kees Roelandschap on X, notes that the latest documents from the United Nations Economic Commission for Europe (UNECE) center on amendments to UNECE regulation 171. These changes were formally adopted into UNECE during the WP29 World Forum in March 2025.

Now, the amendment that will enable SIM will come into force on September 26, 2025. This six-month period after approval is standard UNECE procedure and allows objections from party states that would halt the implementation. Objections to UNECE amendments are genuinely rare, with 95% of amendments passing without objection once the World Forum adopts them.

What This Means for FSD in Europe

System-initiated maneuvers will allow your Tesla to perform actions autonomously, such as changing lanes on a highway, while the driver remains fully responsible for supervision. This is a substantial step up from current regulatory standards that only allow for suggested maneuvers or require explicit driver initiation for every automated maneuver.

This inches towards the more normal “hands-off, eyes-on” approach that Tesla has taken with FSD Supervised in North America and China. This regulatory change will help to provide the legal frameworks needed to deploy more advanced autonomy capabilities in Europe, at least matching what’s available on highways in North America.

Tesla is waiting for EU approval (sigh)
— gorklon rust (@elonmusk) May 12, 2025

Highways Only

The UNECE regulation will apply to all countries that adopt its standards, including the European Union, Japan, and South Korea - unless they specifically block it. This is a relatively positive development, but there are some hiccups.

This regulation only allows for system-initiated maneuvers on highways, not low-speed roadways. That means the city streets portion of FSD’s capabilities - including handling ‘Start FSD from Park` and reaching your destination’s parking spot - still won’t be available in Europe.

In addition, the UNECE framework has stricter requirements for driver monitoring and attentiveness - which means that some nags, including the dreaded wheel nag, are likely to remain a key portion of the experience.

Progress on autonomous driving regulations in Europe has been fraught with indecision and caution, which has been a source of frustration for many who are watching the everyday progress of FSD in North America, and more recently, in China. Even with individual countries recently approving testing of FSD on public roads, harmonized UNECE regulations are the key to wider, consistent deployment, which will allow everyday customers to use it as well.

While the path to full parity with North America may be long and involve even more regulatory machinations, the upcoming implementation of System-Initiated Maneuvers is a big step forward for Tesla owners in Europe. It is a key piece of the puzzle that will help Tesla bring some more features of FSD to Europe, assuming the September timeline holds.