Tesla’s April update described Optimus lines designed for one million robots a year at Fremont and 10 million long-term at Gigafactory Texas. Its more recent Q2 update lists both sites simply as “Construction” and reports no Optimus output. That gap is the key reality check for anyone reading huge robot-capacity headlines.

Big capacity numbers are easy to repeat. They are much harder to interpret correctly.

Tesla’s April Q1 update said its planned first-generation Optimus line at Fremont was designed for one million robots a year. It also said the second-generation line planned for Gigafactory Texas was designed for long-term annual capacity of 10 million robots.

Those are striking figures.

They are also plans, not reported production results.

Tesla’s more recent Q2 update provides the current reality check. In the company’s manufacturing table, both California Optimus and Texas Optimus are marked “Construction.” Neither location has an installed-capacity number. Neither has a production number. Neither has an Optimus delivery or deployment count.

Tesla did provide a useful operational detail: Fremont’s initial Optimus builds are intended for Optimus Academy, where the company says robots will collect training data and support further functionality development.

That is a sensible early-stage use for prototypes and initial production units. It is also very different from operating a mature line at the one-million-unit level.

Think of a house plan.

A blueprint may show a large finished home with many rooms. It tells you what the builder intends to create. It does not tell you whether the foundation is finished, the wiring works, or anyone is living inside.

Capacity claims work the same way.

A line “designed for” one million robots annually is a statement about intended engineering and possible future throughput. Actual output depends on tooling, parts suppliers, factory uptime, quality checks, cost, staffing, and demand. For Optimus, it also depends on whether the robot can perform useful tasks reliably enough to justify scaling.

Tesla’s Q2 update acknowledges that uncertainty in its own general manufacturing language. It says installed capacity is not the same as current production rate, and that production rates depend on many factors, including equipment uptime, component supply, downtime, and regulatory considerations.

That caution is especially relevant here because Tesla has not yet published an installed Optimus capacity figure.

Electrek’s August 20 report said Tesla told JPMorgan that long-term capacity ambitions remained roughly one million at Fremont and 10 million in Texas, while external sales could start as early as the second half of 2027. Those are forward-looking Tesla claims reported through an analyst note, not current output data.

The clean way to follow Optimus is to track four separate numbers:

  • Designed capacity: what a future line is intended to make.
  • Installed capacity: what equipment is actually in place.
  • Production output: how many robots are made in a period.
  • Useful deployment: how many robots complete real work reliably.

Tesla has discussed the first category and confirmed current construction work. It has not publicly supplied the last three numbers in its Q2 update.

That does not make the factory plans unimportant. It simply means that capacity headlines should not be treated as robot-production headlines.

The next meaningful Optimus disclosure would be a current output number, followed by a clear description of what those robots are doing and how reliably they are doing it.

Draft Word Count: 507

Bottom Line

Tesla's stated Optimus capacity figures remain forward plans rather than evidence of achieved production output, so readers should separate targets from demonstrated results.

Sources