Early this morning Tesla released v10 of their FSD Beta as expected. The beta is version 2021.24.15. This release of the Beta is said to include a completely retrained Neural Net, although the release notes for the update are the same as version 9.2 of the FSD Beta. Elon seems very pleased with this specific version, calling it 'next-level' on Twitter.
Lucky participants were able to download the beta this morning and try it out. So far early impressions are that v10 is a noticeable improvement over the previous version. It is now able to navigate areas it just wasn't able to before. Tesla has been making impressive improvements with each FSD Beta release. Lets not forget this beta comes just shy of four weeks after the previous beta.
In this update, the car is more confident in taking left turns at intersections that don't have traffic signals. It accelerates more briskly and generally drives more like a human driver.
Version 10 was able to navigate the famous Lombard Street in San Francisco for the first time without any human intervention. Check out how it performed below:
FSD on Lombard Street
UI Improvements
Elon mentioned that there would also be some UI improvements in this beta. We're now seeing clearer lane markings. When the car approaches an intersection, the visualization will now zoom out slightly to give you a better view of the entire intersection and surrounding cars.
For Model S and X owners, the visualizations on the instrument cluster now also take up the entire screen. Any additional information you add to the IC, like an analog clock, will now be placed on top of the visualizations.
Future Improvements
Elon said this morning that this beta uses all the new vision processing and neural net improvement only for city streets. Highway driving still relies on the current production Navigate on Autopilot codebase. Elon went on to say that Tesla will merge highway and city driving into a single stack that should be included in version 10.1 of the FSD Beta. Over time we should see big improvements in the way Autopilot navigates highways as well.
Version 10 doesn't drive entirely like a human being yet. There was a time when it didn't take the right-of-way when it should have done so. There was a time when it didn't move far enough to the left to pass a pedestrian adult and child in the road with sufficient distance. And it didn't automatically reroute itself when it came to a place where the road was closed.
So v10 is not the ultimate solution for full self driving, but it's getting there. And maybe it's good enough for Elon to give us the "BUTTON" in a few weeks.
Highway stack is still production version, as it is more polished than FSD for now. Single stack for all should release in 10.1.
This version of the FSD Beta saw an increase in the number of testers. We don't know how many, but there were additional Tesla owners invited to test out this version of the FSD Beta. This must mean that Tesla is feeling more confident and wants to slowly start increasing the numbers of testers.
Tesla’s Optimus X account recently dropped another video, following up on its previous dance routines. Previously, Optimus demonstrated impressive footwork, along with a level of agility and balance that has improved immensely since Optimus was first unveiled.
In a new video posted to X, Optimus is seen doing actual work. While it’s still in the early days, Tesla shows off how the robot is learning to perform small tasks.
The real focus of the video isn’t on the tasks Optimus is performing but on how it’s learning to perform these tasks. Optimus is learning by watching humans perform the work through a first-person view.
The video demonstrates Optimus tackling daily tasks that require a good degree of dexterity and environmental awareness, including vacuuming, cooking, and even taking out the trash. While the previous demonstrations were more of a showcase of Optimus’ physical capabilities, this one is more about the learning process.
The key breakthrough, which was emphasized by members of Tesla’s Optimus AI team, is the robot’s new ability to learn complex, bi-manual tasks largely from watching videos of humans performing them. This results in a reduction in direct teleoperation, where real humans control Optimus remotely to generate training data.
Murtaza Dalal, a member of Tesla’s Optimus AI team, shared his excitement about the development - and noted that while teleoperation doesn’t scale, training by watching video does scale — extremely fast.
First-Person Perspective
The ability to learn from human video (while currently focused on a first-person perspective) means Optimus can bootstrap new tasks much faster than before and with less human interaction required. Milan Kovac, an Optimus Engineering team member, elaborated on this, stating:
“We recently had a significant breakthrough along that journey, and can now transfer a big chunk of the learning directly from human videos to the bots... Many new skills are emerging through this process, are called for via natural language (voice/text), and are run by a single neural network on the bot (multi-tasking).”
What’s Next for Optimus?
The road doesn’t stop here. The Optimus team is already looking to see where they can push the barriers of learning, and is looking at having Optimus learn directly from internet videos of humans doing tasks. This means learning from third-person views rather than the more optimized first-person views that it is currently learning from today.
Once Optimus can translate the third-person perspective to the first-person, much like humans do, then it opens up a whole new world of training data. Optimus will be able to learn simply by watching the world around him.
All this rapid advancement is critical for Tesla. Elon has doubled down on Optimus being Tesla’s key to unlocking the biggest market in the world, and it is key to pushing Tesla’s new vision of sustainable abundance.
This is just the first step in a long process - we’re still years away from Optimus making its way to customer homes, but the journey has to start somewhere. For now, it’ll be working hard in Tesla’s labs and factories, and we expect to hear more about Optimus later this year once the production line is up and running.
A concept of what the UI in the Robotaxi may look like
@dkrasniy
According to a report from Bloomberg (paywall), Tesla is planning to launch its Robotaxi network in Austin on June 12th. Bloomberg’s source noted that the date was still in flux, but this is the date Tesla is being planned around.
While the Robotaxi network may launch on June 12th, it’ll be exclusive to Tesla employees initially, allowing Tesla to iron out any issues.
The taxi network isn’t expected to open up to the public until late June at the earliest. During Tesla’s Q1 2025 Earnings Call, Elon stated that the Robotaxi service would only be available to the public in late June or early July.
June 12th is actually much sooner than we expected, so this is an excellent sign for Tesla, who typically pushes back events for finishing touches. The Robotaxi event, for example, was originally scheduled for August 8th but was rescheduled for October 10th.
The NHTSA recently reached out to Tesla, inquiring about how they’d handle certain scenarios, such as adverse weather conditions. This is a good sign that Tesla is satisfying those requirements.
Testing Happening Now
According to Bloomberg’s report, Tesla has already been operating test vehicles using FSD Unsupervised without a safety driver this week. A test engineer was in the passenger seat but did not intervene or provide any input.
Tesla has already been doing preliminary testing with employees in Austin and Los Angeles, but those have been with relatively small internal groups for the most part.
Tesla plans to launch the service with approximately 10 Model Y taxi vehicles initially, providing an opportunity to gather data and user feedback. According to Elon, in the weeks following the launch, Tesla will expand the program to more vehicles, with it hitting several thousand by late 2025.
Tesla recently started the Early Access Program for FSD users in Texas. It’ll be interesting to see how Tesla leverages these users to gather more data or help expand the program when Tesla is finally ready to provide FSD Unsupervised to vehicle owners.
Reducing Variables
There are a ton of moving parts involved in launching an autonomous taxi network, including insurance, cleaning, charging, and more. There’s no doubt Tesla is eliminating as many variables as possible, such as automated charging. While the Robotaxi will support wireless charging, these Model Ys are expected to be charged manually since Tesla hasn’t built the “snake” charger, and Model Ys don’t support wireless charging.
Tesla already has at least some Robotaxi code built into the Tesla app, so it seems like they’re lining up all the pieces.
While Tesla has been relatively focused on California and Texas, we’ve also spotted some Robotaxi mules making their way around Boston and New Hampshire, meaning that Tesla is already thinking about expansion or at least finding out what it’ll take.
We may just be a couple of months away from seeing a Robotaxi revolution in some American cities. While expanding to additional cities with varying weather conditions may be slow, Tesla will be out there proving what they’ve been promising since 2016.
![Tesla’s Optimus Is Now Learning by Watching — First Person Today, Third Person Tomorrow [VIDEO]](https://www.notateslaapp.com/img/containers/article_images/2025/optimus_cleaning.png/bd519f7f31d9c3e7e25c2736ee284efb/optimus_cleaning.jpg)
