According to Elon Tesla may release FSD Beta 10.12 with several key improvements this week.
The last major beta, version 10.11 started
going out in early March, but most testers didn't receive it until v10.11.2, which
was released in April.
We hope this beta will go out a little quicker than the previous one, but it's certainly shaping up to be an
exciting release.
Updated Vehicle Models
FSD Beta 10.12 will contain new, more
detailed vehicle models, at least for its sedan model, but it may include updated models for all the
various vehicle types.
The current sedan visualization is fairly abstract. It doesn't have wheels or many details. The visualization is
modeled after Tesla's key fob for the Model S.
FSD Beta can recognize far more things than it can display on the screen. However, the visualizations are an
important way of how the vehicle communicates with us on what it sees and understands. So with Beta 10.12 Tesla
is including a more detailed sedan model that has wheels and doors.
Although FSD Beta has been able to detect open doors for a while now, the model will now visually show us if any
nearby cars have open doors by highlighting the door in yellow.
Improvements to Unprotected Left Turns
Unprotected left turns have been a key focus over several betas and we're apparently going to see further
improvements in 10.12.
Crossing over multiple lanes when turning left can be intimidating, even for some human drivers. Tesla has
been making continuous improvements to make unprotected left turns more efficient and human-like.
For example, the car will now sometimes start inching slowly, anticipating the last vehicle to pass so that it
can complete the turn promptly and be out of the way of any further traffic.
According to Elon, FSD Beta 10.12 will specifically improve "tricky" unprotected left turns.
Chuck Cook on YouTube does a fantastic job covering some of these left turns. Below you can see how the latest
FSD Beta does taking a left turn onto a primary street with a divider.
Heavy Traffic
In Beta 10.12, we're also expecting to see improvements in heavy traffic. I haven't seen too many issues with the
beta in traffic, except that sometimes the car has a tough time differentiating between a parked vehicle and a
vehicle that's just waiting.
I've encountered situations where the beta tries to go around a car that is stopped due to a traffic light or
traffic and the beta waits for just a few seconds before trying to go around the vehicle.
Hopefully, this is one of the areas that Elon is talking about when he refers to improvements in heavy traffic.
Single Stack
Elon also mentions that Tesla is making good progress on single stack. Single stack refers to a single set of
technologies that will be used for both highway and street driving.
FSD beta is great, but once you get on the highway, you're right back in the old production version.
FSD Beta is far from perfect, but driving on city streets is a completed task and the beta actually does quite
well trying to figure things out.
When we start looking at Autopilot on the highway and some of the issues it still has, like bouncing between lane
markings or a sudden attempt to center itself in a lane that has become wider, those issues are practically
non-existent in city driving.
So while single stack won't be included in beta 10.12, it's good to know that Tesla continues to make progress.
When Tesla is finally able to complete their single stack software we should see huge improvements in highway
Autopilot use.
Release Date
The last FSD beta started going out more than a month ago, so a lot of users are definitely itching for an
update. Elon said earlier this week that beta 10.12 is "probably" going to wide release this week.
The beta could be in QA testing now, but it unlikely that it has been passed on to employees yet as release notes
usually get leaked when that happens.
Hopefully, some of us will be greeted by that sweet notification this weekend, prompting us to install the latest
beta.
Update: Elon tweeted today, Friday, May 6th that there have been "many upgrades to core code, so taking longer to debug issues. Probably Wed/Thurs release." So it looks like we're still a week out from a public release of beta 10.12.
Based On New Build
Lately, FSD betas have been a little behind the times. The latest beta is 2022.4.5.21,
which is roughly two major versions behind. That means that FSD Beta testers still don't have seat heaters in
the launcher, Dog Mode in the Tesla app, browser improvements, vehicle preconditioning improvements and more.
Most non-FSD Teslas are now on a 2022.12 release and 2022.16 is expected shortly.
While it's unlikely that the beta 10.12 will be based on a brand new upcoming build like 2022.16, it's almost
certainly going to be based on 2022.12, which
will please a lot of testers.
It appears that Tesla is as cautious as ever with beta releases. Lately, it has taken several revisions of a beta
before Tesla releases it to everyone.
Tesla says they're now at 100k beta testers in the US and Canada, so they're right to be cautious, but it's not
easy waiting.
Tesla launched two FSD updates simultaneously on Saturday night, and what’s most interesting is that they arrived on the same software version. We’ll dig into that a little later, but for now, there’s good news for everyone. For Hardware 3 owners, FSD V12.6.1 is launching to all vehicles, including the Model 3 and Model Y. For AI4 owners, FSD V13.2.4 is launching, starting with the Cybertruck.
FSD V13.2.4
A new V13 build is now rolling out to the Cybertruck and is expected to arrive for the rest of the AI4 fleet soon. However, this build seems to be focused on bug fixes. There are no changes to the release notes for the Cybertruck with this release, and it’s unlikely to feature any changes when it arrives on other vehicles.
FSD V12.6.1 builds upon V12.6, which is the latest FSD version for HW3 vehicles. While FSD V12.6 was only released for the redesigned Model S and Model X with HW3, FSD V12.6.1 is adding support for the Model 3 and Model Y.
While this is only a bug-fix release for users coming from FSD V12.6, it includes massive improvements for anyone coming from an older FSD version. Two of the biggest changes are the new end-to-end highway stack that now utilizes FSD V12 for highway driving and a redesigned controller that allows FSD to drive “V13” smooth.
It also adds speed profiles, earlier lane changes, and more. You can read our in-depth look at all the changes in FSD V12.6.
Same Update, Multiple FSD Builds
What’s interesting about this software version is that it “includes" two FSD updates, V12.6.1 for HW3 and V13.2.4 for HW4 vehicles. While this is interesting, it’s less special when you understand what’s happening under the hood.
The vehicle’s firmware and Autopilot firmware are actually completely separate. While a vehicle downloading a firmware update may look like a singular process, it’s actually performing several functions during this period. First, it downloads the vehicle’s firmware. Upon unpacking the update, it’s instructed which Autopilot/FSD firmware should be downloaded.
While the FSD firmware is separate, the vehicle can’t download any FSD update. The FSD version is hard-coded in the vehicle’s firmware that was just downloaded. This helps Tesla keep the infotainment and Autopilot firmware tightly coupled, leading to fewer issues.
What we’re seeing here is that HW3 vehicles are being told to download one FSD version, while HW4 vehicles are being told to download a different version.
While this is the first time Tesla has had two FSD versions tied to the same vehicle software version, the process hasn’t actually changed, and what we’re seeing won’t lead to faster FSD updates or the ability to download FSD separately. What we’re seeing is the direct result of the divergence of HW3 and HW4.
While HW3/4 remained basically on the same FSD version until recently, it is now necessary to deploy different versions for the two platforms. We expect this to be the norm going forward, where HW3 will be on a much different version of FSD than HW4. While each update may not include two different FSD versions going forward, we may see it occasionally, depending on which features Autopilot is dependent on.
Thanks to Greentheonly for helping us understand what happened with this release and for the insight into Tesla’s processes.
At the 2025 Consumer Electronics Show, Nvidia showed off its new consumer graphics cards, home-scale compute machines, and commercial AI offerings. One of these offerings included the new Nvidia Cosmos training system.
Nvidia is a close partner of Tesla - in fact, they produce and supply the GPUs that Tesla uses to train FSD - the H100s and soon-to-be H200s, located at the new Cortex Supercomputing Cluster at Giga Texas. Nvidia will also challenge Tesla’s lead in developing and deploying synthetic training data for an autonomous driving system - something Tesla is already doing.
However, this is far more important for other manufacturers. We’re going to take a look at what Nvidia is offering and how it compares to what Tesla is already doing. We’ve done a few deep dives into how Tesla’s FSD works, how Tesla streamlines FSD, and, more recently, how they optimize FSD. If you want to get familiar with a bit of the lingo and the background knowledge, we recommend reading those articles before continuing, but we’ll do our best to explain how all this synthetic data works.
Nvidia Cosmos
Nvidia’s Cosmos is a generative AI model created to accelerate the development of physical AI systems, including robots and autonomous vehicles. Remember - Tesla’s FSD is also the same software that powers their humanoid robot, Optimus. Nvidia is aiming to tackle physical, real-world deployments of AI anywhere from your home, your street, or your workplace, just like Tesla.
Cosmos is a physics-aware engine that learns from real-world video and builds simulated video inputs. It tokenizes data to help AI systems learn quicker, all based on the video that is input into the system. Sound familiar? That’s exactly how FSD learns as well.
Cosmos also has the capability to do sensor-fused simulations. That means it can take multiple input sources - video, LiDAR, audio, or whatever else the user intends, and fuse them together into a single-world simulation for your AI model to learn from. This helps train, test, and validate autonomous vehicle behavior in a safe, synthetic format while also providing a massive breadth of data.
Data Scaling
Of course, Cosmos itself still requires video input - the more video you feed it, the more simulations it can generate and run. Data scaling is a necessity for AI applications, as you’ll need to feed it an infinite amount of data to build an infinite amount of scenarios for it to train itself on.
Synthetic data also has a problem - is it real? Can it predict real-world situations? In early 2024, Elon Musk commented on this problem, noting that data scales infinitely both in the real world and in simulated data. A better way to gather testing data is through real-world data. After all, no AI can predict the real world just yet - in fact, that’s an excellent quantum computing problem that the brightest minds are working on.
Yun-Ta Tsai, an engineer at Tesla’s AI team, also mentioned that writing code or generating scenarios doesn’t cover what even the wildest AI hallucinations might come up with. There are lots of optical phenomena and real-world situations that don’t necessarily make sense in the rigid training sets that AI would develop, so real-world data is absolutely essential to build a system that can actually train a useful real-world AI.
Tesla has billions of miles of real-world video that can be used for training, according to Tesla’s Social Media Team Lead Viv. This much data is essential because even today, FSD encounters “edge cases” that can confuse it, slow it down, or render it incapable of continuing, throwing up the dreaded red hands telling the user to take over.
Cosmos was trained on approximately 20 million hours of footage, including human activities like walking and manipulating objects. On the other hand, Tesla’s fleet gathers approximately 2,380 recorded minutes of real-world video per minute. Every 140 hours - just shy of 6 days - Tesla’s fleet gathers 20 million hours of footage. That was a little bit of back-of-the-napkin math, calculated at 60 mph as the average speed.
Generative Worlds
Both Tesla’s FSD and Nvidia’s Cosmos can generate highly realistic, physics-based worlds. These worlds are life-like environments and simulate the movement of people and traffic and the real-life position of obstacles and objects, including curbs, fences, buildings, and other objects.
Tesla uses a combination of real-world data and synthetic data, but the combination of data is heavily weighted to real-world data. Meanwhile, companies who use Cosmos will be weighting their data heavily towards synthetically created situations, drastically limiting what kind of cases they may see in their training datasets.
As such, while generative worlds may be useful to validate an AI quickly, we would argue that these worlds aren’t as useful as real-world data to do the training of an AI.
Overall, Cosmos is an exciting step - others are clearly following in Tesla’s footsteps, but they’re extremely far behind in real-world data. Tesla has built a massive first-mover advantage in AI and autonomy, and others are now playing catch-up.
We’re excited to see how Tesla’s future deployment of its Dojo Supercomputer for Data Labelling adds to its pre-existing lead, and how Cortex will be able to expand, as well as what competitors are going to be bringing to the table. After all, competition breeds innovation - and that’s how Tesla innovated in the EV space to begin with.
