Tesla has reached another milestone for its autonomous vehicle technology, bringing Full Self-Driving (FSD) Unsupervised to Giga Berlin.
The German factory is now the third, following Fremont and Giga Texas, to deploy driverless Tesla vehicles internally—moving newly built Model Y SUVs autonomously from the end of the production line to the outbound logistics lot without any human intervention.
The use of FSD Unsupervised was spotted in a drone flyover of Giga Berlin by pilot Tobias Lindh, clearly showing the Model Ys moving to the outbound lot and navigating to their designated shipping lane without a driver behind the wheel.
Tesla first rolled out FSD Unsupervised at its Fremont factory earlier this year. There, production vehicles began autonomously driving a 1.2-mile (1.9 km) route from the assembly line to the logistics area. The system uses a vision-based neural network trained on real-world driving data, allowing the cars to navigate factory grounds—complete with traffic, pedestrians, and moving equipment—entirely without human input.
Shortly afterward, Giga Texas followed suit. In Austin, newly built Model Ys traverse a 1.4-mile (2.25 km) route, while Cybertrucks take a shorter yet more complex 0.6-mile path that includes a steep 17% grade and tunnel passage constructed by The Boring Company.
Since then, the two U.S. factories have collectively logged over 50,000 driverless miles (84,600km), helping Tesla refine FSD under conditions that closely mimic public roads but allow for safer data collection and monitoring.
Although the specific route details at Giga Berlin have not been released, the deployment of FSD in Germany suggests that Tesla is continuing to build a robust foundation for future expansion of autonomous technology across global operations.
While Tesla’s Unsupervised FSD is currently limited to internal factory logistics, it represents a meaningful precursor to broader applications. The company is preparing to launch a paid robotaxi network in Austin in just a few weeks. These factory deployments provide a controlled yet unpredictable environment to test and perfect the software ahead of real-world use on public roads.
