Report that Tesla Telling Suppliers to Support 600 Teslabots Per Week by End of 2025

by

There is report that Tesla has given suppliers guidance on Optimus production: 600 units per week by the end of 2025. Ten weeks at the rate of 600 Optimus bots per week would by 6000 units.

8 thoughts on “Report that Tesla Telling Suppliers to Support 600 Teslabots Per Week by End of 2025”

  1. Musk is clearly a fan of hardware centric iterative development. Look at Spacex rocket development. IF NASA is the sniper making a billion calculations and testing to support its aiming and only then shoots, Musk first shoots with a necessary minimum of prep and then aims again based on where the previous bullet went. Look at FSD with Tesla; he has a millionish drivers doing his testing and frequently updating the algorithms based on their experiences, a very interative way to product improvement. If Optimus can be developed to be a universally useful and flexible robot, Musk is likely to make it happen. What he might not be saying is that we can inhabit Mars, but really a very few humans and many Optimus robots are the right mix of population.

  3. Goldfinch – A video from a few months back had it walking up stairs.
    Sure, this year they will “only” make 3-6 thousand, but then next year with 50,000 is when things really kick into high gear.
    You are clearly a butt hurt leftist, who has been told to hate Elon, and so you grabbed your pitchfork and hit the streets.

    • OK, so it can gradually ascend stairs. It might be able to run around the factory walls like Fred Astaire in a virtual environment, but the real world has dropped screws and loose packing tape. For a company with a value similar to TSMC to be dependant of the production of 6000 units of unknown quality is odd.

      Its curious that you surmise my opinion of Elon Musk with how dubious I am about the market cap. Musk alone mostly explains the numbers, not the tech.

      • Agreed. And the edge cases where a humanoid robot is both better than the existing plethora of autonomous robot assembly machines custom made for specific jobs AND better than humans at repetitive high precision work that costs more for human labor…are vanishingly small, if they exist at all.
        Similarly, I can’t think of a single thing that a Teslabot could do to make it worth storing & charging in my 900sq NYC apartment. I’ve seen robots fold laundry and put it in a basket on a table, but going to the common laundry room with a cart using the elevators, paying for the machine, knowing which clothes to wash hot or cold, putting them back in the laundry basket/bag and restocking them in my closets and drawers is a whole other level of complexity, and I or my wife do while daydreaming of more important things too!
        I know there’s a shortage of caregivers and other high time/low pay workers, but there’s even greater complexity for those, plus a certain amount of controlled strength to help move a human carefully. Would you trust your elderly parent to a Teslabot?

  4. You both seem to be lacking the insight on the vitrual training environments and the scale of compute being implemented by Tesla to support this. The hardware is capable of doing a lot more than what the current slow motion demonstrations are performing. The ball catching is a guide as to the axial velocity. I was developing an autonomous machine, but the training was impractical to achieve without virtual training. The virtual training environments are what is more key to what the hardware ends up performing.
    I’ll bet that a Teslabot can far outperform Joe I won’t pardon my son Biden.

  5. Joe Biden can just about walk upstairs.
    A Tesla bot cannot walk upstairs,
    I still would not want a clone army of Joe Bidens in my factory.
    Those mere 6000 Teslabots might sell as novelty items, but considering that Musk is saying that Tesla is pivoting to be a robotics company, and its cars (which people are losing interest in buying) are less important, its not looking good…

  6. Estimations are overblown. Realistically, perhaps some production will happen, but the volume is the question. Best guess is they will be used internally in factories to test them—not just a few, but in larger numbers.

    In some of the latest videos, they were tele-operated. So not true autonomy here. Perhaps they did it to avoid embarrassment if something went wrong in real time, I don’t know.

    They know the limits of current hardware and software. Software can be updated, but hardware is harder to modify.

    The plus side is they should be less complex and faster to build than cars. Lots of people will be willing to spend $10k+ on a robot if the hardware is good enough, and later software updates make it very useful.

Comments are closed.