Connect with us

News

Elon Musk shares Tesla FSD Beta 9.2’s behind-the-scenes technical release notes

Credit: @brandonee916/Twitter

Published

on

Tesla released FSD Beta 9.2 over the weekend, and while the update’s official Release Notes were identical to FSD Beta 9.2, there were a lot of improvements that were rolled out by the company under the hood. These improvements were so notable that Elon Musk stated on Twitter that 9.2’s release had to be moved back a bit to address some last-minute issues. 

The Tesla CEO then decided to share the technical release notes for FSD Beta 9.2, which explains every optimization that has been introduced in the recent update. Based on what Musk has shared, it appears that FSD Beta 9.2 is a massive step up from V9.1, allowing the advanced driver-assist system to navigate through inner-city streets in a safer and more confident manner. 

The following are the technical release notes for FSD Beta 9.2, as shared by CEO Elon Musk on Twitter. 

  • Clear-to-go boost through turns on minor-to-major roads (plan to expand to all roads in V9.3).
  • Improved peek behavior where we are smarter about when to go around the lead vehicle by reasoning about the causes for lead vehicle being slow.
  • V1 of the Multi-modal prediction for where other vehicles expected to drive. This is only partially consumed for now.
  • New Lanes network with 50k more clips (almost double) from the new auto-labeling pipeline.
  • New VRU velocity model with 12% improvement to velocity and better VRU clear-to-go performance. This is the first model trained with “Quantization-Aware-Training”, an improved technique to mitigate int8 quantization.
  • Enabled Inter-SoC synchronous compute scheduling between vision and vector space processes. Planner in the loop is happening in v10.
  • Shadow mode for new crossing/merging targets network which will help improve VRU control. 

Apart from the technical details of FSD Beta 9.2’s release, Elon Musk also hinted at the upcoming updates coming for the advanced driver-assist system. The CEO noted that FSD Beta 9.2 would be followed by V9.3 and V9.4, but after this point, Tesla may already release FSD Beta 10. V10 would feature “significant architecture changes,” as per the CEO. Musk did not explain further, though speculations suggest that FSD Beta 10 would be the update that would transition Tesla into a single FSD stack. 

Lastly, Musk provided some updates on the release date of the FSD Beta Download Button, which has experienced substantial delays. The CEO explained that the “Button” would probably roll out with Beta 10 or V10.1, which, based on the current release cycles of FSD Beta’s updates, would likely be in the next couple of months, barring any unexpected delays.  

“Beta 10 or maybe 10.1. Going to pure vision set us back initially. Vision plus (coarse) radar had us trapped in a local maximum, like a level cap. Pure vision requires fairly advanced real-world AI, but that’s how our whole road system is designed to work: NN’s with vision,” Musk explained. 

Don’t hesitate to contact us with news tips. Just send a message to tips@teslarati.com to give us a heads up. 

Advertisement
-

Simon is an experienced automotive reporter with a passion for electric cars and clean energy. Fascinated by the world envisioned by Elon Musk, he hopes to make it to Mars (at least as a tourist) someday. For stories or tips--or even to just say a simple hello--send a message to his email, simon@teslarati.com or his handle on X, @ResidentSponge.

Advertisement
Comments

Elon Musk

SpaceX turned a heralding moment for Starship into its greatest moment

Starship reached orbit despite losing an engine, deployed 26 Starlink V3 satellites on Flight 14.

Published

on

By

SpaceX’s Starship reached orbit for the first time on Monday, and for a few nail-biting minutes it looked like it wouldn’t. During ascent on Flight 14, one of Ship 41’s six Raptor engines shut down early, and SpaceX’s livestream host Dan Huot told viewers the team had decided not to commit to orbit. Minutes later, after what Huot described as a lot of conversation in the control room, the final poll came back in favor, and a roughly 19 second burn of a single Raptor pushed the ship into orbit about 170 miles up.

The reversal matters because SpaceX had written the exit ramp into the mission plan. The company said it would only fire the orbital insertion burn if flight controllers confirmed enough backup hardware remained for the deorbit burn, a condition Teslarati laid out ahead of the flight. Losing an engine was exactly the scenario that rule was built for.

Pressing forward fits Elon Musk’s history. Falcon 1 failed three straight times before its fourth launch reached orbit in 2008, with SpaceX nearly out of money, and Starship was developed by flying prototypes until they broke. What changed this year SpaceX going public, and with $SPCX sliding below its IPO price in July when Flight 13 slipped, the short interest climbed significantly, as Teslarati reported at the time. A Starship potentially lost today with revenue generating next-gen Starlink satellites aboard would have landed directly on shareholders.

That pressure showed up after orbit. SpaceX cut a flight planned to last nearly 10 hours to about three, moving splashdown from west of Chile to the North Pacific near Hawaii. SpaceX gave no reason, though Musk said this month the company was being extremely cautious about debris risk. The single Raptor for deorbit worked, and Ship 41 completed its flip and landing burn before breaking apart in the water, an outcome SpaceX expected. Musk has structured SpaceX’s governance to shield long term bets from market pressure.

The payload is the bigger business story. Musk posted that all 26 Starlink V3 satellites deployed and are “operating nominally.” Each V3 is rated for about 1 Tbps of downlink and 160 Gbps of uplink, so this single launch adds roughly 26 Tbps, about 10 times what a Falcon 9 load of V2 Mini satellites adds. The V3 is too large for Falcon 9, making Starship the only vehicle that can build out the planned 100,000 satellite constellation, at up to 60 per flight once it reaches routine service. Unlike the 20 V3 units on Flight 13, which reentered on a suborbital path, these will raise their orbits and could begin serving customers within weeks and bring in hundreds of millions of additional dollars in projected Starlink revenue.

SpaceX has already begun winding down Falcon 9 Starlink launches from Florida in favor of Starship. Reported targets put Flight 15 as early as October 19, leaving about three weeks to diagnose Monday’s engine shutdown before the next orbital attempt.

Continue Reading

News

Tesla Cybercab fleet doubles to well over 100 units

Published

on

(Credit: Teslarati)

Tesla quietly doubled the size of its Cybercab fleet within the Robotaxi program in Austin, Texas, over the weekend to well over 100 units.

The move not only establishes more of the steering-wheel-less and pedal-less vehicles within the ride-sharing fleet Tesla has been operating for a year, but it also solidifies a more robust Robotaxi fleet as a whole.

Riders started receiving notifications from the Robotaxi app that stated: “Cybercab fleet has doubled: more rides available.”

Tesla first launched rides in the Cybercab in early September, although the Robotaxi fleet has been active for over a year, as rides began last Summer. Cybercab is truly Tesla’s most crucial vehicle release yet, as it is the first car any company has built that is geared toward full-fledged and end-to-end autonomy, never needing human intervention for anything.

Only available in Austin at the current time, Cybercab has two seats and has been spotted testing around various U.S. states and regions; Tesla plans to deploy the Cybercab in various U.S. cities in the coming months as a best-case scenario.

Tesla Cybercab gets initial tie-in to localized, in-house cathode plant

The availability of the Cybercab has doubled from just 58 units last Monday to 125 the following Friday. Marking a substantial increase in Cybercab availability, the additional ride-sharing units are more than welcome, as wait times for Cybercabs, especially, were quite high.

The dramatic increase is a sign that demand for Robotaxi is growing and Tesla is feeling more confident that its driverless ride-hailing suite, especially its Full Self-Driving software, is able to handle any traffic situation without explicit direction or supervision from a human being.

Continue Reading

News

Tesla has a ‘no human contact’ approach for Semi production

Published

on

Tesla is advancing a fully automated pipeline for the 4680 battery cells used in its all-electric Semi, spanning production from Giga Texas through shipment and direct consumption on the line at the new dedicated Semi Factory in Sparks, Nevada.

The approach was outlined by Tesla at its September 24 Semi Handover event, which launched high-volume production at its new 1.8-million-square-foot plant in Nevada, which sits adjacent to Gigafactory Nevada and is designed for an annual production rate of 50,000 trucks per year.

After years of pilot builds and what was a four-year-long redesign of the truck, Tesla moved the Semi from 2170 batteries to its in-house 4680 cells, which are made in Austin. The change cuts battery mass and total energy while holding range, a key step in making volume production a realistic possibility.

Cells will leave Giga Texas in trailers, and at the Nevada Semi plant, Tesla intends for a dedicated line to unload those trailers automatically, station the cells, and feed them straight into pack and vehicle assembly.

Both Lars Moravy, Tesla’s VP of Vehicle Engineering, and Dan Priestley, the Head of Tesla’s Semi program, described the goal as a “zero human touch point” from the moment the trailer arrives in Texas until a finished Semi drives off the production line in Nevada.

The unloading system that Moravy and Priestley described is just one piece of a much broader automation push. The plant uses what Tesla calls the highest-capacity electric monorail conveyance in vehicle manufacturing, carrying frames-in-white simultaneously. Powder-coating replaces conventional paint, and many processes that would normally require operators have been designed out.

Tesla has repeatedly said that “the best part is no part,” and the cell-handling plan extends that philosophy from the cell factory floor in Texas all the way to final assembly in Nevada.

If executed as described, the closed-loop flow would reduce labor, handling damage, and inventory buffers while tightening quality control on a component that represents a large share of the truck’s cost and weight. It also shortens the physical and organizational distance between two factories separated by more than 1,200 miles. The Semi itself now shares a bar-wound stator and other components with the Cybertruck, further linking Tesla’s passenger and commercial production systems.

High-volume output is expected to ramp gradually after the first trucks left the new line in April 2026. Early customers include PepsiCo, DHL, and U.S. Foods. Whether the automated trailer-to-line process reaches the promised zero-touch standard will be visible in the coming months as production scales. For Tesla, the Semi factory is another test of how far it can push “the machine that builds the machine” across sites.

Continue Reading