Tesla FSD v15 is starting to roll out, bringing a massive 25X improvement in handling safety-critical situations and major upgrades across its entire end-to-end driving network
Tesla is upgrading nearly every part of its end-to-end driving network, and this is a huge upgrade
Here's everything new:
• Safety - Upgraded vision encoder shared across FSD, Robotaxi and Optimus, trained on millions of internet videos for better generalization and data efficiency, with a claimed 25X improvement in critical intervention rate
• Tesla Reality View - A completely new real-time generative driving visualization that shows what FSD is planning to do. Choose fullscreen, maps, split view or quiet mode. It replaces separate model signals with a single visualization while using less compute
• Smarter navigation - Learns route preferences from Tesla's fleet, starting with the beginning and end of trips. Future v15 releases will expand this to the entire route
• Personalized parking - Learns your parking preferences and displays multiple parking options at your destination
• Improved driver monitoring - Better attention detection when wearing hats or sunglasses, with less strict monitoring in higher-speed driver profiles
• Better detection - Improved recognition of road signs, small animals, road debris, chains, potholes and deep standing water
• Smarter speed handling - More natural speeds across driving profiles, school zones, speed limits and bad weather. FSD now slows for speed cameras
• Bike Rack Mode - Enables FSD when bike racks or similar hitch accessories are installed
• Improved Autoshift - Uses FSD-based gear predictions for better gear selection
What makes this particularly interesting is the upgraded vision encoder shared across FSD, Robotaxi and Optimus
Tesla is bringing its vehicles and humanoid robots closer together through a shared foundation for understanding the world
And FSD v15 is just beginning its rollout
Tesla is upgrading nearly every part of its end-to-end driving network, and this is a huge upgrade
Here's everything new:
• Safety - Upgraded vision encoder shared across FSD, Robotaxi and Optimus, trained on millions of internet videos for better generalization and data efficiency, with a claimed 25X improvement in critical intervention rate
• Tesla Reality View - A completely new real-time generative driving visualization that shows what FSD is planning to do. Choose fullscreen, maps, split view or quiet mode. It replaces separate model signals with a single visualization while using less compute
• Smarter navigation - Learns route preferences from Tesla's fleet, starting with the beginning and end of trips. Future v15 releases will expand this to the entire route
• Personalized parking - Learns your parking preferences and displays multiple parking options at your destination
• Improved driver monitoring - Better attention detection when wearing hats or sunglasses, with less strict monitoring in higher-speed driver profiles
• Better detection - Improved recognition of road signs, small animals, road debris, chains, potholes and deep standing water
• Smarter speed handling - More natural speeds across driving profiles, school zones, speed limits and bad weather. FSD now slows for speed cameras
• Bike Rack Mode - Enables FSD when bike racks or similar hitch accessories are installed
• Improved Autoshift - Uses FSD-based gear predictions for better gear selection
What makes this particularly interesting is the upgraded vision encoder shared across FSD, Robotaxi and Optimus
Tesla is bringing its vehicles and humanoid robots closer together through a shared foundation for understanding the world
And FSD v15 is just beginning its rollout



