Tesla converts Fremont car line to build 20,000 Optimus robots a week

Tesla has begun assembling Optimus robots on a high-volume line inside its Fremont factory, claiming the output meets the standards of a commercial product. The company stated in its third-quarter shareholder letter that the current manufacturing process is suitable for a real-world product, marking a shift from prototype development to mass production.

The assembly line occupies space previously used to build Model S and Model X vehicles. Tesla cleared the old vehicle line in 46 days to make room for a robot production capacity designed for one million units annually. Elon Musk told investors on a Wednesday earnings call that most equipment for that capacity is already installed in Fremont, with weekly output expected to rise toward 20,000 units as the line ramps up.

Musk described the production ramp as an S-curve with a shallow initial slope. The engineering team is currently refining the robot’s AI, electromechanical design, reliability, and service life. Tesla plans to complete this work by the end of next year, at which point it will begin delivering Optimus to outside companies for tasks that are dangerous, repetitive, or require around-the-clock operation.

A central challenge in scaling robot intelligence is the lack of a massive fleet to generate training data, a resource Tesla’s cars possess. Musk noted that there is no vast fleet of Optimus units to gather data from yet. The company’s solution is an internal Optimus Academy, where 10,000 to 20,000 robots will practice various tasks. These robots will feed their learned behaviors into the next model version, effectively using every unit from the Fremont line as a source of training data.

Hardware upgrades are also scheduled for next year, including the integration of Tesla’s AI5 chip into the robots. Musk said a new video demonstrating the robot completing tasks driven by AI rather than preprogrammed motions should be released within one or two months. He criticized rival humanoid robot demonstrations for relying on scripted routines that may not hold up in real-world conditions.