Tesla converts Fremont factory line to build 20,000 robots weekly

Tesla has begun assembling its Optimus humanoid robots on a high-volume line at its Fremont, California factory. The company stated in its third quarter shareholder letter that the units leaving this line are the first to meet production quality standards. Tesla confirmed that the manufacturing process is now suitable for commercial products.

The production space previously housed the Model S and Model X vehicle lines. Tesla cleared the old car assembly equipment in 46 days to make room for robot production. The facility is designed to handle one million units per year. Most of the necessary equipment is already installed in the plant.

Elon Musk outlined the production ramp on the recent earnings call. He expects weekly output to climb toward 20,000 robots as the line reaches full speed. Musk described the growth trajectory as a shallow S curve in its initial phase. The engineering team is currently refining the robot's AI, electromechanical design, and reliability. Tesla aims to finish this work by the end of next year. The company plans to begin delivering Optimus units to outside companies for dangerous and repetitive tasks at that time.

A new training method addresses the lack of a large fleet for data collection. Tesla cars generate AI training data from millions of vehicles on the road. Optimus does not have this advantage yet. The company introduced the Optimus Academy, a program where 10,000 to 20,000 robots will practice various tasks. These robots will feed their learned behaviors into the next version of the AI model. Every unit produced at Fremont serves as a data source for this system.

Hardware upgrades are also scheduled for next year. Tesla plans to integrate its AI5 chip into the Optimus platform. Musk said a new video demonstrating AI-driven task completion will be released within one or two months. He contrasted this with rival humanoid demos that rely on scripted routines. He stated that those scripted performances would not hold up in real-world conditions.