With FSD V13.2.6 continuing to make its way to AI4 vehicles, Tesla has been on a streak with minor FSD improvements since the launch of FSD V13 just a little over two months ago.
FSD V13 brought a new slate of features, including Start FSD from Park, Reverse, and Park at Destination. It also introduced full-resolution video input using the AI4 cameras at 36hz and made use of the new Cortex supercomputer to get faster and more accurate decision-making.
So, what’s next with FSD V14? Tesla gave us a sneak peek at what’s next for FSD.
FSD V14
The standout feature of FSD V14 will be auto-regressive transformers. While that’s a complex term for those unfamiliar with AI or machine learning, we’ll break it down.
Auto-Regressive
An auto-regressive transformer processes sequential data in time, using that information to predict future elements based on previous ones. Imagine completing a sentence: You use the words already written to guess what comes next. This process isn't just about filling in the blank; it's about understanding the flow of the sentence and anticipating the speaker's intent.
FSD could analyze a sequence of camera images to identify pedestrians and predict their likely path based on their current movement and surrounding context. The system's auto-regressive nature allows it to learn from past sequences and improve its predictions over time, adapting to different driving scenarios.
Today, FSD reacts to what it sees, but soon it’ll be able to anticipate what will help, much like humans.
Transformers
The second part of that term is transformer, which is a component used to understand the relationships of elements inside a time sequence. It identifies which parts of the input are most crucial for making accurate predictions, allowing the system to prioritize information much like a human would. Think of it as weighing different pieces of evidence to arrive at a conclusion. For example, a transformer might recognize that a blinking turn signal is more important than the color of the car when predicting a lane change.
Putting It Together
Putting all that together, Tesla’s use of auto-regressive transformers means they’ll be working on how FSD can predict the plans and paths of the world around it. This will improve FSD’s already powerful perception and allow it to predict how other vehicles and vulnerable road users (VRUs) will behave.
What it all comes down to is that FSD will be able to make better decisions and plan its paths by making more informed, human-like decisions. That will be a big step towards improving V13 - which already has some very effective decision-making.
Larger Model and Context Size
Ashok Elluswamy (Tesla’s VP of AI) stated that FSD V14 will see larger model and context sizes in FSD V14, which coincidentally are listed in the upcoming improvements section of FSD V13.2.6. If we compare what Ashok said to what’s listed in the upcoming features section, the model and context sizes should grow by 3x.
Interestingly, Ashok says that AI4’s memory limits context size. Context is essentially the history of what the vehicle remembers, which is used for future decisions. Since this information is stored in memory, it’ll always be limited by memory, but it’s worth noting that Ashok mentioned that Tesla is restricted by the memory in the AI4 computer.
Leverage Audio Input
Tesla is already gathering audio data in existing FSD versions so that it can start training models with audio as well, truly making FSD more human-like. According to Ashok, FSD V14 will be the first version to take advantage of audio input for FSD driving. This will primarily be used for detecting emergency vehicles, but we can see this expanding to other sounds that help humans adjust their driving, such as car crashes, loud noises, honking, etc. At the very least, FSD could be more cautious when hearing a noise that matches an accident or vehicle honking.
FSD V14 Release Date
We haven’t heard from Elon Musk or Ashok Elluswamy about when FSD V14 will arrive. Ashok previously stated that FSD V13.4 would see audio inputs being used, but at Tesla’s earnings call, Tesla said that audio input would become relevant in V14, making it seem like Tesla may scrap V13.4 for V14.
Since Tesla is planning to launch their Robotaxi network in Texas this June, which is just four months away, FSD V14 may be the version used for its autonomous taxi fleet.
Subscribe
Subscribe to our newsletter to stay up to date on the latest Tesla news, upcoming features and software updates.
Following the first major service area expansion, Tesla is already pushing another update to the Robotaxi app and service as a whole. The new version includes more quality-of-life improvements, and most importantly, a big change to Robotaxi’s pickup logic that makes the entire experience much smoother.
Robotaxi App Update
This latest update is another one focused on polishing the user experience before Tesla adds more vehicles to the fleet and more riders to the network. The most significant change here is that Robotaxi will now navigate more dynamically. Instead of strictly travelling to the designated pickup point, the vehicle will stop where you are or navigate directly to you if you are nearby, making the pickup just that much easier.
Tesla also introduced a small batch of improvements to improve the app, which is still only available for early testers. First up is a vehicle finder arrow - which is shown in the Robotaxi app to help you find your approaching or parked Robotaxi.
This should be a big help in crowded situations, as visually distinguishing a Model Y Robotaxi from a Robotaxi Model Y is nearly impossible, barring the small Robotaxi logos and lack of a driver.
Tesla has also added in automatic dark/light mode switching to match system settings and some other minor design tweaks and bug fixes. It seems that Tesla is close to their final iterations on the Robotaxi app and will hopefully add it to the official app stores soon. However, an Android version is still MIA, and using the Robotaxi itself is still also limited to invitees, so Tesla may be waiting for those to be available before adding the app to the Apple App Store.
The full release notes for this 25.7.6 app update are listed as:
I am pointing the way! Find your vehicle with an arrow
Automatic light/dark mode setting to follow the system
Design tweaks
Bug fixes
This will be huge and make summon even better.
I didn't get to fully test it, but Robotaxi no longer strictly navigates to the pickup point. I noticed it stops right where I am if I am ahead of the pickup point.
The most exciting part of the update isn’t the piece for Robotaxi - but what it implies for Smart Summon instead. The new dynamic pickup logic could be a huge improvement to how Summon works. At the end of the day, it is exceptionally similar - homing in on the user or a targeted location, and then parking nearby.
Summon currently requires a line of sight, and the range is exceptionally limited, meaning you can’t use it in most larger parking lots without having to walk out halfway to meet the vehicle. It’s also slow enough that by the time you get your vehicle rolling - even with Summon Standby active - you’ll probably have walked most of the way there yourself.
A true come-to-me Summon feature that can navigate both city streets and parking lots is definitely the goalpost for customer vehicles - one that doesn’t need to be directly supervised. Robotaxi is the first real-world deployment of the same technology that future builds of Smart Summon will eventually run on.
Sometime in the near future, you’ll be able to Summon your Tesla directly to you - having it navigate the streets from wherever it found parking. All the learnings from Robotaxi are going to come fast - and they’ll all be amazing to finally get on customer cars.
After months of no updates, Elon Musk has announced that Tesla will soon release a new FSD update, in what he’s calling a “step change improvement,” so expectations will be high for this one. The biggest change is that Tesla is merging in the changes they’ve been making for Robotaxi FSD into the consumer software branch.
Elon also provided an update on the rollout of Robotaxi to additional US cities like San Francisco, as well as FSD for other regions like Europe and China. Let’s dig in - as this is some of the most exciting FSD-related news we’ve seen in a while.
However, the Tesla AI team isn’t standing still. They’ve been making constant improvements to the Robotaxi version of FSD, so hearing that Tesla will integrate that version into FSD Supervised is certainly exciting.
Musk says that Tesla still needs to validate that the improvements they’ve made for Austin in particular don’t cause regressions elsewhere. Regression testing is a necessary step when training neural nets, as they can learn and unlearn things extremely rapidly, as one new behavior can cause a regression in another.
We need to validate that improvements for Austin don’t cause regressions elsewhere.
For Europe and China, we are awaiting regulatory approval. Hopefully, soon🤞
Essentially, we’re at a good spot here. The fact that Tesla is now working on merging Robotaxi FSD into FSD Supervised is a good sign, even if there’s work to be done. We expect these builds to go out to the FSD testing fleet (not the Robotaxi validation testers we’ve seen throughout Texas). These dedicated testers will make sure there are no regressions - and once they give the thumbs up, the standard rollout process begins. Given that the Robotaxis are running HW4 and Tesla is merging Robotaxi FSD into FSD Supervised, we expect this to be an HW4-only rollout.
Once validation is complete, Tesla will likely roll out the update to employees, followed by early access influencers (and potentially also Early Access Program members). Once it is in their hands and we see good progress there, it’ll begin rolling out to customers en masse.
Elon commented on X last night, saying they’re looking good for Q3 release, which puts us in the September timeframe.
Global Rollout Status
Elon also provided an update on two other items - launching Robotaxi in other cities, and launching FSD into other markets - and made it clear that the next hurdle isn’t on Tesla, it is on regulatory approval. Bureaucratic red tape seems to be the biggest obstacle to Tesla’s autonomy ambitions lately - but it is progressing well in some regions, such as Australia.
For the Bay Area, as well as potentially other cities that Tesla is considering, such as Phoenix, Arizona, approval is pending from regulators. Elon mentioned that regulators were being reasonable, which isn’t something we hear often.
As soon as the regulators approve (they are being quite reasonable tbh)
For other markets, including China (for HW3 vehicles) and Europe, the red tape is still in place. Both of these markets have big regulatory hurdles that Tesla needs to overcome. In China’s case, every new update must be validated with the government, which takes additional time. In Europe’s case, we’re still awaiting the green light from UNECE regulators to approve autonomous vehicles on European roads.
Elon provided a much-needed update on FSD and Robotaxi, one that we hadn’t had in a long time. Current users can look forward to what is likely a major capability boost, sourced from Robotaxi in Austin, while those outside of North America now have some additional hope that FSD may arrive before the end of 2025.
![Tesla Hollywood Diner Opens: A Look at the Food, Merch, Features and More [VIDEOS]](https://www.notateslaapp.com/img/containers/article_images/2025/tesla-diner-mosaic.webp/1f15f0d02c11a96d7fa1399f47581304/tesla-diner-mosaic.jpg)
![First Look at Tesla’s 2025.26 Update [Photos + Videos]](https://www.notateslaapp.com/img/containers/article_images/2025/grok-app.webp/f66c0424fde99ab9adb2b041d854f723/grok-app.jpg)
![Tesla Diner: First Look at Its Food and Potential Collectibles [VIDEO]](https://www.notateslaapp.com/img/containers/article_images/2025/tesla-diner-logo.webp/feaab9ffb76b08874e474c8abdf2684d/tesla-diner-logo.jpg){kind=logo}
