The factory floor that built the Model S for over a decade is now being rebuilt to manufacture humanoid robots. Elon Musk confirmed in early July that installation of Tesla's first Optimus production line is underway at the Fremont, California plant, posting a photo of himself with the production team and cautioning that initial output will be extremely slow. It is the clearest signal yet that Tesla's most ambitious product since the electric car is moving from prototype theater to industrial reality.
From Model S to Optimus in 46 Days
Tesla announced the end of Model S and Model X production during its Q4 2025 earnings call, and the final vehicles rolled off the Fremont line in early May. What happened next was quintessentially Tesla: according to reporting from Yahoo Finance and Teslarati, the company tore down the legacy vehicle line in just 46 days, ripping out concrete trenches and robotic equipment to make way for humanoid assembly.
Tesla's VP of Vehicle Engineering, Lars Moravy, has described the new line as modular in design, revealing that full Optimus production will ultimately involve some 40 lines because of the sheer number of actuators and the intricate assemblies required for the robot's limbs and torso. That complexity underscores a point Musk has made repeatedly this year: Optimus contains roughly 10,000 unique parts, nearly all of which lack a mature supply chain.
A Deliberately Slow Start
On the company's Q1 2026 earnings call, covered by Electrek, Musk said Optimus production would begin at Fremont in late July or August — just months after the last Model S left the building — while warning that it was practically impossible to predict this year's production rate for an entirely new product on an entirely new line.
The July 1 photo of Musk walking the Fremont line, first reported by TrendForce and Benzinga, came with the same caveat: output will be extremely slow at first. Supply chain analysts have sketched an aggressive ramp scenario — low hundreds of units per week by late summer, potentially climbing toward 1,000 or more per week by autumn, with some insiders floating year-end capacity as high as 2,500 units weekly. Those figures come from supply-chain sources rather than Tesla itself, and they should be treated with caution given the company's track record on robot timelines.
Key facts about the Fremont ramp:
- The first-generation line has a designed annual capacity of 1 million Optimus units.
- Robots built in 2026 are earmarked for internal factory testing and data collection, not commercial sale.
- There are currently no public sales, pre-orders, or waitlists; consumer availability is not expected before late 2027 at the earliest.
- A second, dedicated Optimus factory is under construction at Giga Texas, expected to come online around 2027 and eventually build a higher-volume Gen 4 variant.
A Humbling History of Targets
The Fremont milestone arrives against a backdrop of missed projections. In January 2025, Musk predicted Tesla would build roughly 10,000 Optimus robots that year. The actual figure fell dramatically short, and Musk conceded in early 2026 that no Optimus units were yet doing genuinely useful work in Tesla's factories.
That candor appears to have reshaped the company's messaging. Rather than headline-grabbing unit targets, Tesla is now emphasizing supply chain maturation, internal deployment, and iteration speed. Investors will get their next substantive update at the Q2 earnings call on July 22, where confirmation of the production start date — and any revision to it — will be closely watched.
Why It Matters
Tesla's Fremont conversion is the most consequential bet yet that humanoid robots can be manufactured like cars: on automated lines, at automotive volumes, with automotive cost curves. No competitor — not Figure AI with its BotQ facility, not China's fast-scaling Unitree — has publicly committed to anything close to a million-unit annual capacity for a single line.
If Tesla makes the economics work, the labor implications reach far beyond its own gigafactories, touching warehousing, manufacturing, and eventually the home. If it stumbles, the failure will validate skeptics who argue humanoids remain a decade from commercial maturity. Either outcome will define the competitive landscape, because rivals from Silicon Valley to Shenzhen are calibrating their own production plans against Tesla's timeline. Unitree's IPO prospectus even names the Optimus ramp as a direct competitive threat — a remarkable acknowledgment that the robot wars have moved from the lab to the factory floor.
For now, the story is one of concrete, rebar, and modular tooling in a repurposed corner of Fremont. The robots will come slowly at first. Whether they come in the millions is the trillion-dollar question of the decade.
