If I were to hallucinate what it is and why it's worded that way: AI robot space is in need of a hyper-realistic game engine with better physics than Unity/Unreal style non-deformable rigid body mechanics, that's also way faster than 1x completely unlike engineering FEM sims, and this cater to that need
No, the "action" part is the distinction. Their world model is conditioned on robot actions for example, which gives you two things the video gen alone can't: predict the future frames that follow a given action (change the action, get a different future from the same starting frame), and run it in reverse to infer the actions behind observed frames or output the actions needed to hit a goal (the output is motor commands abd not video frames).
> Cosmos 3 Nano is the compact version with 16B parameters and optimized for efficient inference. It’s designed to run on workstation-grade compute, like the NVIDIA RTX PRO 6000 GPU for real-time robotics inference and physical AI applications.
Looking forward to trying this out on my $10000+ workstation grade GPU that I need an equally expensive set up to run.
Most of the examples they've chosen seem.. not good? What an odd mix of bad game engine and AI slop. I can't imagine that this stuff makes good training data for real-world applications.
These demos honestly look pretty good to me. But it is objectively true that this and similar technologies are used at huge scale by every leading autonomous vehicle manufacturer, so we can inductively reason that it _is_ good enough for that use-case. I don't work on Cosmos, but I am currently working on a superficially similar non-open technology at Nvidia used by many of these leaders which, in my opinion, produces similar quality. Some of the open research for it is here:
This release unifies those capabilities with a Mixture-of-Transformers (MoT) architecture built around two towers.
Reasoner tower: A vision-language model (VLM) ... This serves as the ‘brain’ that reasons about the world before any generation happens.
Generator tower: Generates future observations and action sequences. This tower uses a diffusion-based process to generate physics-aware video and action outputs that are conditioned on the reasoner tower’s understanding.
This sort of approach (and others i've seen like it) always appeal to my inner engineer, trying to optimize and balance tradeoffs between model architectures and combine two things to yield the best of both worlds
But based on my understanding of the Bitter Lesson (http://www.incompleteideas.net/IncIdeas/BitterLesson.html), this is precisely the wrong approach in the long term. I'm linking the actual text of the bitter lesson because I think it's misunderstood (or I just don't agree with how i've seen it used in discourse). Specifically:
The bitter lesson is based on the historical observations that 1) AI researchers have often tried to build knowledge into their agents, 2) this always helps in the short term, and is personally satisfying to the researcher, but 3) in the long run it plateaus and even inhibits further progress, and 4) breakthrough progress eventually arrives by an opposing approach based on scaling computation by search and learning. The eventual success is tinged with bitterness, and often incompletely digested, because it is success over a favored, human-centric approach.
This architecture feels specifically like "trying to build knowlege into the agent that will help in the short term" but will plateau long term. That's not to say that there won't be some interesting learnings or things built on top of it, but I doubt that there's a lot of juice to squeeze with this kind of approach IMO.
13 comments
[ 2.8 ms ] story [ 19.9 ms ] threadStill impressive nonetheless given its artificially generated training sets.
Beats nano banana 1 but not yet competitive with 2 or seedance2, grok imagine,etc.
> Generates future observations and action sequences.
Is that just a complicated way of saying video gen?
Looking forward to trying this out on my $10000+ workstation grade GPU that I need an equally expensive set up to run.
https://github.com/nv-tlabs/3dgrut/
https://github.com/NVIDIA/harmonizer
https://github.com/NVIDIA/instant-nurec
https://github.com/nvidia/ncore
Nvidia also is integrating Gsplat into at least what I work on and contributing upstream.
https://github.com/nerfstudio-project/gsplat
But based on my understanding of the Bitter Lesson (http://www.incompleteideas.net/IncIdeas/BitterLesson.html), this is precisely the wrong approach in the long term. I'm linking the actual text of the bitter lesson because I think it's misunderstood (or I just don't agree with how i've seen it used in discourse). Specifically:
This architecture feels specifically like "trying to build knowlege into the agent that will help in the short term" but will plateau long term. That's not to say that there won't be some interesting learnings or things built on top of it, but I doubt that there's a lot of juice to squeeze with this kind of approach IMO.