FSD v14.3.9 Release Notes (2026.27.6)

This is the first public release of FSD v14.3.9. The employee version included slightly different release notes for Teslas new Automatic Collision Avoidance feature.

Automatic Collision Evasion

Automatic Collision Evasion can activate Full Self-Driving (Supervised) to help keep the vehicle safe and then continue driving. It may engage while you are driving manually if it detects that you are not sufficiently attentive to the road (for example, reaching toward the back seat), or if Full Self-Driving (Supervised) was unintentionally disengaged.

See your Owners Manual for more details.

FSD (Supervised) v14.3.9

Full Self-Driving (Supervised) v14.3.9 includes:

The Reinforcement Learning (RL) stage used to train the FSD neural network has been upgraded, improving performance across a wide range of driving scenarios.
The neural network vision encoder has been upgraded, improving understanding in rare and low-visibility situations, strengthening 3D geometry understanding, and expanding traffic sign comprehension.
The AI compiler and runtime were rewritten from the ground up with MLIR, delivering a 20% faster reaction time and speeding up model iteration.
Unnecessary lane biasing and minor tailgating behaviors have been mitigated.
Parking spot selection and maneuvering are more decisive.
Parking location pin prediction is improved and now appears on the map with a P icon.
Responses to emergency vehicles, school buses, right-of-way violators, and other rare vehicles are enhanced.
Handling of small animals is improved by focusing RL training on harder examples and adding rewards for better proactive safety.
Traffic light handling is improved at complex intersections with compound lights, curved roads, and yellow light stopping, driven by training on hard RL examples sourced from the Tesla fleet.
Handling of rare and unusual objects that extend, hang, or lean into the vehicle path is improved by sourcing infrequent events from the fleet.
Handling of temporary system degradations is improved by maintaining control and automatically recovering without driver intervention, reducing unnecessary disengagements.
Driver monitoring sensitivity is improved with better eye gaze tracking, eye wear handling, and higher accuracy in variable lighting conditions.
The model is unified across Actually Smart Summon, FSD, and Robotaxi for more capable and reliable behavior. (New in FSD v14.3.2 for non-Cybertrucks, New in FSD v14.3.4 for the Cybertruck)
Actually Smart Summon max speed is now increased to 8mph (13km/h). (Added in FSD v14.3.3, not available on Cybertruck)
You can help Tesla improve Self-Driving by selecting an intervention reason on the main screen after taking over. (New in FSD v14.3.2)
You can now see distance traveled in FSD (Supervised) without an intervention. The Self-Driving App will also show your longest intervention-free streak. (Added in FSD v14.3.3)
Parking options now display on the map when arriving at your destination. (Added in FSD v14.3.4)

Upcoming Improvements

Expand reasoning to all behaviors beyond destination handling.
Add pothole avoidance.

Full Self-Driving (Supervised)

With your active supervision, Full Self-Driving (Supervised) can drive your Tesla almost anywhere. It can start from a parked position, change lanes, choose forks to follow your navigation route, navigate around vehicles and objects, make left and right turns, and park at your destination. You and anyone you authorize must remain attentive and use additional caution. It does not make your vehicle autonomous. Do not become complacent.

Full Self-Driving (Supervised) is enabled on your vehicle. To use it, press the Start Self-Driving button on the UI or press the right scroll wheel button once. You can disable Full Self-Driving (Supervised) in Autopilot Settings.

FSD (Supervised) v14.2 Lite

FSD (Supervised) v14.2 Lite includes:

Distillation of intelligence from HW4 V14 into HW3, allowing HW3 to learn how to handle scenarios using HW4 V14 as a guide. This unlocks HW4 improvements for HW3, including Reinforcement Learning (RL) and offline models.
Improved proactive and reactive responsiveness across categories such as navigation handling, merges and forks, pedestrian interactions, traffic lights, and vehicle cut-in scenarios.
Improved comfort in nominal scenarios with fewer false slowdowns, smoother steering, and more consistent lane centering.
Introduced parking, unparking, and reversing capabilities.
Added Arrival Options so you can choose where FSD should pull over or park: in a Parking Lot, on the Street, or in a Driveway.
Speed Profiles are now available at all times to further customize driving style preference.

Start Self-Driving from Park

A button appears on the screen when driver requirements are met. Driver requirements include (but are not limited to):

Driver is seated in drivers seat
Driver is buckled
Cabin Camera is unobstructed

Brake Confirm requires the driver to press and release the brake pedal before FSD (Supervised) can be activated. When Brake Confirm is disabled, pressing the button allows Self-Driving to apply the brakes and engage FSD (Supervised). If driver detection requirements are not met or if FSD (Supervised) does not detect a driver, you will be required to press and release the brake pedal to confirm. You can enable Brake Confirm in Self-Driving > Brake Confirm.

This feature does not make your vehicle autonomous, and you must remain attentive and be ready to take over at any time. Only a licensed driver authorized to operate the vehicle should be allowed access to the drivers seat position any time the vehicle is not in Theater or Arcade modes.

Arrival Options

You can select an arrival option such as Parking Lot, Street, Driveway, and Pull Over for Robotaxi-style drop offs.
Your preferences for arrival options and preferred parking positions are saved for each destination.

Speed Profiles

FSD (Supervised) now determines the appropriate speed based on a mix of driver profile, speed limit, and surrounding traffic.

Introduced a new Speed Profile, SLOTH, which uses lower speeds and more conservative lane selection than CHILL.
Driver profile has a stronger impact on behavior. The more assertive the profile, the higher the max speed.
The right scroll-wheel up/down now adjusts the Speed Profile setting rather than your precise max speed offset selection in mph/kph.