13 comments

  • vessenes8 hours ago
    Really interesting. Upshot: a well trained multimodal video generation model has a world representation model trained inside it. They’ve done some work lifting this world model out and deploying it to robots, where it seems to work well.<p>On the one hand, this isn’t a new idea, and the quality video models certainly have understanding of materials, light, the world (at least in an Occam’s razor sense of understanding). I’m not aware of a video lab that’s turned itself into a robot lab yet, though; perhaps this would be a first, or a new sort of obvious-in-retrospect business path: train video model, sell video generation, scale, use scale to train robot things: profit.<p>I found their hands very interesting - looks like a bunch of stuff hidden in gloves - Xiami’s Robotics-1 foundation model just released videos of users training on some pretty standard looking grippers; to the point that there are demo videos of people putting on gripper type gloves to make video to train that model.<p>The BFL model looks like it doesn’t need that at all. Given the difficulty of the hardware side, I’ll be curious to see what they do with this.
    • mike_hearn6 hours ago
      I think this tactic is standard, no? Nvidia has demoed robots being trained using video generation models, and Waymo has been doing that for a while too.<p><a href="https:&#x2F;&#x2F;waymo.com&#x2F;blog&#x2F;2026&#x2F;02&#x2F;the-waymo-world-model-a-new-frontier-for-autonomous-driving-simulation&#x2F;" rel="nofollow">https:&#x2F;&#x2F;waymo.com&#x2F;blog&#x2F;2026&#x2F;02&#x2F;the-waymo-world-model-a-new-f...</a>
    • lairv5 hours ago
      FYI several other video labs are getting into robotics<p>Luma Labs <a href="https:&#x2F;&#x2F;lumalabs.ai&#x2F;news&#x2F;luma-open-physical-ai-lab" rel="nofollow">https:&#x2F;&#x2F;lumalabs.ai&#x2F;news&#x2F;luma-open-physical-ai-lab</a><p>Runway <a href="https:&#x2F;&#x2F;runway.com&#x2F;product&#x2F;robotics" rel="nofollow">https:&#x2F;&#x2F;runway.com&#x2F;product&#x2F;robotics</a>
  • GiffertonThe3rd8 hours ago
    The video at around 3.30min, where the robot arm took 3 attempts to reseat the window trim, was quite unnerving - I have not seen such resolving before. Is it new or am I way out of the loop?
    • dinfinity7 hours ago
      You are indeed out of the loop. Google did something arguably more impressive more than a year ago, with a VLA based bot replacing a tensioned timing belt: <a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=2AAFiuEP7iE" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=2AAFiuEP7iE</a>
    • quadrature4 hours ago
      if you want to catch up, i would say Generalist is the state of the art in dexterous manipulation tasks, you can learn more here <a href="https:&#x2F;&#x2F;generalistai.com&#x2F;blog&#x2F;gen-1" rel="nofollow">https:&#x2F;&#x2F;generalistai.com&#x2F;blog&#x2F;gen-1</a>
    • ra8 hours ago
      Yes that was impressive.
  • altern86 hours ago
    This is awesome.<p>What&#x27;s sad to me is that we have all this awesome technology, but movies are worse than ever.<p>I usually watch movies from decades ago just to find something decent and it&#x27;s amazing how with goofy-looking puppets the storytelling was 1000 times better.<p>This is probably an unrelated rant, sorry
    • phoghed6 hours ago
      Survivorship bias, possible that you just aren’t watching all the shitty ones that nobody remembers.
      • altern86 hours ago
        That&#x27;s exactly the ones I like! On Saturdays I watch cheesy horror movies from the 80s.<p>But with my GF I watch whatever is rated &gt;= 8 on Imdb and the 8s from years ago are ALWAYS much, much better than the recent ones.<p>Maybe I just don&#x27;t like the style. All the over explaining, etc.
        • stronglikedan3 hours ago
          I regularly stream movies from Screambox, FilmRise and Alter as background music while I work from home. I love cheesy horror, but they&#x27;re just bad enough to keep me company without distracting me
        • vanviegen5 hours ago
          I think most movies drop a bit in IMDb score over their first few years. That might (partly) explain it. Or you just prefer older movies . :-)
          • altern83 hours ago
            Interesting! I didn&#x27;t know that. Could be both, too :-)
      • throwthrowuknow3 hours ago
        OP replied but I’m chipping in too, I’ve also been watching a lot of older movies specifically the more B list and C list which weren’t successful cult classics and they’re still better than anything new. The low production and low talent C list ones are still cheesy and the performances don’t land anywhere near as well as they could but they feel more genuine and creative than what passes for a movie these days.
      • acdanger6 hours ago
        time is the great filter
        • altern86 hours ago
          What&#x27;s a good recent movie I could watch? Today is Friday
          • Rebelgecko3 hours ago
            Odyssey is supposed to be pretty good. Main complaint I&#x27;ve seen is that it&#x27;s not as horny as the source material.
            • altern83 hours ago
              Thank you! I&#x27;ll take a look
          • phoghed4 hours ago
            Was going to pull out my default recommendation of Oldboy but then I realized it’s 23 years old, which just aged me greatly
            • thomastjeffery2 hours ago
              No Other Choice (by the same director) is incredible, and even relevant to this post.
            • altern83 hours ago
              Welcome to the club! :-)
          • mejutoco3 hours ago
            The sheep detectives is pretty good<p><a href="https:&#x2F;&#x2F;www.imdb.com&#x2F;title&#x2F;tt32565993&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.imdb.com&#x2F;title&#x2F;tt32565993&#x2F;</a>
            • altern83 hours ago
              Thanks! I&#x27;ll take a look
          • thomastjeffery2 hours ago
            Do you like horror? That&#x27;s the only genre where risky projects are getting consistent funding. Obsession and Backrooms are both really good.
    • doctorpangloss2 hours ago
      would you refuse to invest in BFL because Robin Rombach doesn&#x27;t know anything about making movies?<p>sora is done. can you find me one person brave enough to say, &quot;Bill Peebles, a nice guy, doesn&#x27;t know anything about making movies?&quot; this isn&#x27;t just a VC problem, this is why it is called the Hollywood No, but that said: maybe the hollywood no is the problem? in video games it has been called Toxic Positivity. are you getting it now?
  • flufluflufluffy7 hours ago
    Ok the phrasing here, it’s, it’s just -<p>&gt; However, compared to more specialized approaches for representation learning they produce less disentangled representations, which puts a ceiling on their usefulness for tasks that require world understanding.<p>Only an LLM would use a less disentangled representation of the concept “more entangled” when trying to explain to people in the real world why less disentangled representations are not as useful for modeling the real world.
    • phoghed6 hours ago
      We’re just pulling signs of LLM touched writing out of our ass now. Might be time to move on from the accusations, assume all writing is at least LLM assisted and judge it purely on the quality.
      • teodosin3 hours ago
        Yeah, as if people haven&#x27;t used words like &quot;irregardless&quot; way before LLM&#x27;s already
        • cwillu31 minutes ago
          After all, the LLM&#x27;s learned it from <i>somewhere</i>
      • visarga4 hours ago
        I really hope this is how we are going to adapt to LLM era.
    • pennomi7 hours ago
      Bad news, humans write silly things all the time. If anything, LLMs are less likely to make awkward phrasings than people, because they aren’t found in the training data very often.
    • flufluflufluffy2 hours ago
      goddamn people it’s a joke
    • stronglikedan3 hours ago
      &gt; Ok the phrasing here, it’s, it’s just -<p>That dash definitely means an LLM wrote this comment. &#x2F;s
      • cwillu30 minutes ago
        Nah, that was an n-dash, which is widely respected as humans-only.
  • rlupi9 hours ago
    It&#x27;s nice to see partnerships between European startups.
    • DarkNova69 hours ago
      Wasn’t Flux purchased by Meta?
      • xnx9 hours ago
        No. Just a partnership: <a href="https:&#x2F;&#x2F;www.bloomberg.com&#x2F;news&#x2F;articles&#x2F;2025-09-09&#x2F;meta-to-pay-140-million-to-use-black-forest-labs-ai-for-images" rel="nofollow">https:&#x2F;&#x2F;www.bloomberg.com&#x2F;news&#x2F;articles&#x2F;2025-09-09&#x2F;meta-to-p...</a>
  • doubleorseven7 hours ago
    this is the future. but also the present.<p>if you know some one out of the software development&#x2F;engineering worlds, please forward this to them
  • htrp7 hours ago
    OpenAi cries in sora
  • takd8 hours ago
    Amazing work.<p>u.i. Zugló robot mikor?
  • idiotsecant7 hours ago
    We&#x27;re barreling head-on into a crisis where humans have even less value unless they own the means of production. Dark days ahead, friends.
  • tancoai_dev4 hours ago
    [dead]
  • deton39919 hours ago
    [flagged]
  • johnbarron8 hours ago
    If Amazon does not buy them, the CEO should be fired...
    • kensai7 hours ago
      Are you American? We would prefer it to stay a European company. I personally hope Mistral buys them or somehow they unite to make a stand.
      • Tenoke6 hours ago
        Mistral is much worse in its respective field than Flux in their own so I hope not.
        • vanviegen5 hours ago
          Perhaps.. Mistral seems to be focussing more on LLM application rather than foundation models. Seeing that the latter may be becoming commodities, Mistral may come out ahead.
    • Yokohiii5 hours ago
      BFL resides in the same region as BMW and Audi. Quite wealthy and pretty fierce to protect their car industry. If Audi or any other german car maker sees BFL as the future of automation, it will be rather hard to buy them out.
    • windexh8er7 hours ago
      I think you meant to type &quot;promoted&quot;.