Today, we announce our team’s progress in pursuing a different type of foundation model for robotics: the Direct Video Action Model (DVA), which does our best to take robotics and turn it into a generative modeling problem we can scale. Technical blog: rhoda.ai/research/direct-vid…

Rhoda AI checked whether scaling web video pretraining actually helps real robots, using a real customer task. Unpacking 10kg boxes of bearings and sorting the packaging. Turns out yes. Bigger models do better, more pretraining compute does better, and the compute gap is widest when you don't have many robot demos. Refreshing to see a company display this amount of rigour. Rare these days.

We recently came out with our blog post, Does Scaling Web-Video Pre-training Help Real Robots Do Real Work? This is a pretty exciting moment for robotics, because as far as I am aware, it is the first evidence that robot foundation models can scale on internet video data; not merely on teleoperation, UMI data, or egocentric data, but on the data you can truly find anywhere. It’s the necessary requirement for reaching massively powerful models.

An insightful study from @tongzhou_mu and the @RhodaAI team that examines how web video pre-training translates to robot performance. The gist: pre-training doesn’t just help; it scales, i.e. you can continue to reap greater benefits as you put more effort into pre-training. A better pre-trained model will lead to a more robust policy. But how you achieve that pre-training quality is up to you! Whether by scaling the model, or scaling your training tokens, it’s a design choice informed by your compute constraints, training budget, and how much you care about compute optimality.

I was a guest on the @MIT_CSAIL podcast to discuss robots, AI, and what the near future might look like. @klgiven did a great job steering the conversation, and I think much of our chat is accessible to non-experts - hope you find it interesting! bit.ly/4zruZdt Also on Spotify: open.spotify.com/episode/5QX…

I'll be at ICML to present Latent Forcing! Come stop by the poster on Tuesday at 2:00 PM KST, Hall A #3507 Reach out if you want to chat about the future of generative models!

Robot learning is moving beyond policies built for one robot, one scene, one task. At MIT, we’re exploring a different path: turning video world models into embodiment-agnostic robot policies. Introducing VERA: a 14B video-to-action system that controls robots across embodiments, skills, and environments. From zero-shot pick-and-place on a real Panda arm to contact-rich cube reorientation with a 16-DoF robotic hand. Different robots. Different environments. Different tasks. Same video planner. Same weights. We’re open-sourcing everything so you can fine-tune VERA for your own robot setup too. Deep dive in the thread: 🔗 vera.csail.mit.edu 🧵 (1/7)

Delighted to have you on the team! Here’s to many new insights together :)

Introducing MilliVid, our new method for long-context video generation! MilliVid creates videos that are consistent over long time spans, without using retrieval heuristics or 3D maps! (1/n) davidcharatan.com/millivid/#

The shell game is a fun challenge that cannot be solved by looking at a single frame. The model has to track every move, from the moment the object is hidden. Excited to share this!

Here’s something we’ve never seen done before. Real-world tasks are long and ambiguous. Solving them requires visual memory and state tracking. Most robot policies only see the last few frames. Ours doesn't. We put our DVA, FutureVision, to the perfect testbed: the shell game 🐚. The DVA nails it.

1/ We are speed running industrial robotics. It took us just 19 days from the first day of data collection to filming a 2.5-hour continuous run of our model autonomously breaking down industrial containers — zero human intervention. The data efficiency of our DVA model is fundamentally changing how fast we bring robots out of the lab and into the factory. Autonomous operation with 3 hours of data collection at a customer factory.

Everyone is talking about "World Models" for robotics, following the buzz from GTC 2026. But the research landscape is shifting so fast it’s difficult to keep up. In my view, here are the two dominant paradigms currently grounding the video world models in robot control. --- Paradigm 1: Use the Video Model as a Simulator The first major approach is using video world models to simulate reality. In this framework, the model predicts "what happens next" in either pixel space or latent space, conditioned on text prompts or robot actions. Much like traditional analytical simulators (e.g., IsaacSim, MuJoCo, ManiSkill), these learned simulators are used for data synthesis, planning, and evaluation. 1.1 Synthesizing Data for Policy Training A representative work is DreamGen [1]. Given an initial frame and a language instruction, a fine-tuned video model synthesizes clips of a robot completing a task. An inverse dynamics model then labels these videos with actions to train a separate robot policy. GR00T N1 [2] uses a similar strategy. Alternatively, models can act as interactive simulators where agents (like UniSim [4]) or humans (like Interactive World Simulator [3]) generate data through interaction. Key Advantages: Thousands of hours of "synthetic experience" at a lower cost and the ability to safely simulate rare, dangerous edge cases. 1.2 Inference-Time Planning Instead of following a fixed path, robots can use video models to "imagine" multiple future outcomes. In V-JEPA 2 [5], an action-conditioned video model evaluates different action sequences to find the best next step. This "imagination-based planning" is also a core theme in CLASP [6], SWIM [7], VLP [8], GPC [9], DreamDojo [10], and Cosmos Policy [11]. The challenge remains fitting this heavy computation into real-time control budgets. 1.3 Policy Evaluation Video models allow us to test policies before they ever touch physical hardware. Veo Robotics [12] demonstrates that these models can accurately predict relative performance and perform "red teaming" to expose safety violations. This approach is also seen in IRASim [13], 1XWM [14], Ctrl-World [15], and others. Summary of Paradigm 1: While powerful, there is no "free lunch." These methods depend on prediction accuracy. Our physical world is complex, and teaching video models to handle every edge case without hallucinating physics remains a significant challenge. --- Paradigm 2: Use the Video Model as a Policy The second, more integrated paradigm is using the generative video model as the policy (decision-maker) itself. Because the native outputs are videos rather than robot actions, several methods have been developed to obtain control signals. 2.1 Generating Video and Action Jointly A straightforward idea is to add an action decoder to the video model backbone and run video and action denoising jointly during inference. Representative works include DreamZero [16], Cosmos Policy [11], Motus [17], PAD [18], GR-1 [19], and GR-2 [20] (note that the GR series are not diffusion models). This method leverages the rich spatiotemporal priors of pre-trained models with minimal architecture changes. 2.2 Extracting Visual Representations for Action Generation Rather than full generation, many methods use video models to extract deep visual representations to guide action generation. Example works include VPDD [21], VPP [22], UVA [23], UWM [24], Video Policy [25], and DiT4DiT [26]. A major advantage here is that you don’t necessarily need to run multiple denoising steps on giant models, making real-time control easier, though it remains unclear if the full potential of the video models is being utilized. 2.3 Open-loop Video Generation + Video-to-Action Translation A rising trend involves generating a "desired future" video and using a separate inverse dynamics model to translate that video into actions. UniPi [27] pioneered this, followed by This&That [28], TesserAct [29], and 1XWM Self-Learning [30]. Some methods generate videos of humans completing tasks (Dreamitate [31], Gen2Act [32], LVP [33]) and translate those to robot actions. This approach allows video models to do exactly what they were trained for: video generation. 2.4 Closed-loop Video Generation + Video-to-Action Translation Open-loop generation often leads to hallucinations: the model might "see" the robot picking up an apple that isn't actually there. Closed-loop generation avoids this by constantly conditioning on the latest real-world observations, replacing generated frames with real ones in the next call. Recently, mimic-video [34] and LingBot-VA [35] reached real-time speeds using KV caching and partial denoising. Most notably, the DVA [36] model released this month manages real-time generation with full video denoising, which means denoising pure noise all the way to clean video for every step. This approach seems really promising to me, because it reduces robot control into a problem of real-time video generation, which can directly benefit from large-scale video pre-training. --- To me, the key takeaway from this evolution is how we have begun bridging the gap between the digital and physical worlds. Instead of trying to manually program every physical law, we are leveraging the implicit physics embedded in billions of web videos. Whether we use these models as simulators or as direct policies, the objective is the same: providing robots with a “physical common sense.” By reformulating robot control as a challenge of real-time video generation, we may be on the verge of a new scaling law for embodied intelligence. [References in the comment]

Most robot demos are “golden runs”: a perfect take selected from many attempts. But real-world deployment is about Continuous Operation. Watch our DVA model tackle a real-world decanting task for 1.5 hours straight: Uncut, Zero human intervention. 🧵👇

Robot video foundation models can build very powerful robot manipulation policies! These policies enable complex, dexterous manipulation, solve tasks that require long-term visual memory, and do in-context demonstration learning!

Excited to see @rhodaai come out of stealth! As their advisor, I've had a front-row seat of their work on Direct Video-Action Models which reformulates robot control as video generation. The data efficiency here is super promising. Complex industrial tasks learned from just ~10 hours of robot data. Big things ahead!

Vincent has been an inspiration for me since I started in AI — it's not an exaggeration that I wouldn't have done research at all if it were not for him. Thank you for the kind words!