Pinned Tweet
🚀 Excited to share that our paper Fast-ThinkAct has been accepted to #CVPR2026! 🎉
Efficient Vision-Language-Action reasoning via verbalizable latent planning — enabling embodied agents to think fast internally without lengthy textual reasoning.
⚡ Achieves 9.3× faster inference (89% latency reduction) than ThinkAct-7B — bringing Reasoning VLA closer to real-time robotic control.
📄 arxiv.org/abs/2601.09708
🎥 jasper0314-huang.github.io/f…
🙌 Huge congrats to @chipinhxyz, @yunzeman, @ZhidingYu, @CMHungSteven, @jankautz, Yu-Chiang Frank Wang, @FuEnYang1
#EmbodiedAI #PhysicalAI #VLA #Robotics #NVIDIAResearch @NVIDIAAI @NVIDIARobotics
Fu-En (Fred) Yang retweeted
Internship Opportunity 📣
We’re hiring Research Scientist Interns at NVIDIA’s newly established NVIDIA Research Lab in Singapore, Agentic Robot Intelligence Laboratory (ARIEL)!
Our focus is agentic robotics: building the next generation of agentic frameworks for robots.
nvidia.wd5.myworkdayjobs.com…
Fu-En (Fred) Yang retweeted
After working on Dream-1 for the last 4 months, here's my take:
LLMs vs WAMs isn't the right framing. A model like Dream-1 will soon be a tool call for LLMs, and that gets more feasible as LLMs inevitably get faster and cheaper.
Two things we didn't put in the blog post🧵👇
Fu-En (Fred) Yang retweeted
Thanks @_akhaliq for sharing our work!
👇 Try our interactive demo with your own LLM!
🎬 Demo: huggingface.co/spaces/uclanl…
🔗 Website: openvlharness.github.io
📄 Paper: arxiv.org/abs/2610.26860
💻 Code: github.com/OpenVLHarness/Ope…
OpenVLHarness - a Hugging Face Space by uclanlp
Visual tool-use agent on your own model
huggingface.co
OpenVLHarness
Real World Visual Agentic Tool-Use with Multimodal Memory
paper: huggingface.co/papers/2610.2…
Fu-En (Fred) Yang retweeted
🚀 Introducing OpenVLHarness:
Turn any LLM into an evidence-based reasoner that thinks and codes in visual space.
📈 GPT-6 Astra gains +9.9 points on average over 22 datasets.
⚡ Beats Codex and other VL harnesses by 7.7–8.5 points, at ≤37% of their cost.
🧵 More details ↓
Fu-En (Fred) Yang retweeted
NVIDIA Cosmos world models have come to life with real-time interactivity! 🌎🤖🚘
We are releasing our post-training framework for real-time interactive world simulation for Physical AI. The framework builds upon Cosmos 3 through causal training, distillation, and optimized streaming inference. The post-trained models stream visual observations via autoregressive generation in response to incoming actions and scene controls.
We are also releasing example model checkpoints for robotics, autonomous vehicles, and sim-to-real transfer. The release includes optimized streaming inference code to power the models.
Cosmos world foundation models provide broad visual knowledge. Each simulation setting adds its own controls, data, and evaluation requirements. Our shared recipe and example models give researchers and developers a practical starting point for building specialized interactive world simulators.
Huge shoutout to the team for the dedication, creativity, and hard work poured into this release! I am incredibly proud of this work and feel fortunate to lead such a talented team. A special thank-you to the broader Cosmos group for building the world foundation models together. Team green 💚
I am excited to see how you build your own interactive world simulator with Cosmos!
🏠 Project website: research.nvidia.com/labs/cos…
🤗 Model checkpoints: huggingface.co/collections/n…
💻 Code: github.com/NVIDIA/cosmos
Fu-En (Fred) Yang retweeted
Recently, there’s been some "P(doom)" in the robot learning community around the idea that LLMs will solve robotics. I believe LLM agents are great for planning, but handling dynamics and dexterous manipulation still requires a fast System 1—or System 0.
Inspired by @typesafeai JEV’s release of a fast System 1 model, we hope this result offers another perspective on why System 1 still matters—and why there’s still so much worth pursuing in robotics research.
We partnered with @robocurve @chooi_jeq , an independent evaluation company, to benchmark Dream-1. It achieved the highest score among the models evaluated on RoboDojo-RC Tier 1: 59, while being faster and cheaper. ⚡
Regardless of VLA, WAM, LLM, etc, they can all be complementary. Ultimately, I’m glad to see the community keeping an open mind and exploring different ways to solve robotics. Ultimately, the objective matters more than any particular approach: making robots more accessible.
Dream-1 by @dreamlabs_ai is the first model to complete "Stand Up Bottles", a difficult robotics task that frontier LLMs struggle with.
Fu-En (Fred) Yang retweeted
Introducing Flora, an open model suite for scaling robotics data engines.
It watches robot and human video and annotate what happened (timestamped instructions) and how well it went (rewards and failure reasons).
🔗 research.nvidia.com/labs/gea…
We use it to train NVIDIA GR00T. 🧵
Fu-En (Fred) Yang retweeted
Today, Dream Labs is coming out of stealth with Dream-1, a robot foundation model that controls bimanual robots it has never seen, on tasks it was never shown, in places it has never been, from $3/hour.
Dream API is live for everyone, today. 🧵
Fu-En (Fred) Yang retweeted
Today, Dream Labs is out of stealth.
In February 2026, while we were at NVIDIA, @scott_e_reed @youliangtan @GeYunhao and I started calling a new kind of robot model a World Action Model, one that learns how the world moves from video, then uses that to decide how to act.
Today, we’re releasing Dream-1 through a public API that anyone in the world can use.
When you rent a car you’ve never driven, you don’t learn to drive all over again. Within a few minutes, it feels like your own. Robots should work the same way: a good model should be able to pick up a new robot and get to work, without being retrained from scratch.
That’s what we’ve spent the last 6 months on, and it’s what Dream-1 does. It controls bimanual robots with parallel grippers it has never seen before, zero-shot, on unseen tasks in unseen environments.
Generalization has been my North Star in robotics: first by distilling video world models into VLAs (DreamGen), then by learning to act from those world models directly (DreamZero), and now by transferring across robots (Dream-1). I’ve come to believe it’s the only direction that lasts. Pushing a narrow model from 90% to 99.9% on a single task is real work, but much of it gets washed away once a model that generalizes arrives.
Since the start of the company we’ve stayed single-minded about generalization.That focus is what got Dream-1 to control robots it was never trained on, on tasks it was never shown. And it’s what makes a public API possible at all: a model that only does one task on one robot can’t serve millions of users.
Dream-1 is still a research beta, and it will make mistakes. But it already drives cars it’s never seen.
Generalization wins in the end, and the end is closer than it looks.
Here’s Dream-1 taking the wheel of a robot it’s never seen. Full details 👇
Fu-En (Fred) Yang retweeted
Gemini 4 Argon taught a robot to play cornhole just by watching a video.
It mapped the physics, coded its own simulation, and taught @Apptronik's Apollo humanoid to sink 7 throws in a row.
Fu-En (Fred) Yang retweeted
We are open-sourcing V1.0 of our agentic robotics framework, ASPIRE, with full real robot skill discovery and self improvement capabilities! On the day Astra arrived, our lab switched the backend model and saturated all the benchmarks to ~99% over night.
And the frontier is already on the move:
- Opus 5.5 slightly edges out Astra on our robot simulation bench.
- Sonnet 5.5 hits the Pareto frontier of cost vs. robot capability, $12/task.
We first introduced ASPIRE and /robot skill library in June, way before the big models could reliably do robotics. A rising frontier tide lifts all boats, but Astra was a 0->1 tsunami on many tasks.
Now you can also feel the Physical Bitter Lesson at home like we did on that fateful day ;)
links in thread
Fu-En (Fred) Yang retweeted
GPT and Claude now operate real robot arms. What about open-weight models? 🤔
Introducing ORA: Open Robot-Use Agents.
An open harness powered by @deepseek_ai V4.1-Flash, ORA operates robots from visual input—zero-shot, no VLA—and competes with frontier-model agents.
Fu-En (Fred) Yang retweeted
Robotics has spent years scaling demonstration data and model size. What happens when we scale EXPLORATION for robot agents?
We're incredibly excited to share RAX, an agentic robotics system that learns by scaling online exploration.
A coding agent connects directly to the robot’s perception and control APIs, learns from feedback, and builds programs that perceive, plan, and act. The model stays frozen. As exploration scales, the agent improves its programs and builds up reusable skills.
The results:
• ZERO teleoperation or action demonstrations on real robots—from insertion with 0.1 mm clearance and air-fryer manipulation to deformable objects and long-horizon mobile whole-body control.
• +37 percentage points on average over leading learned policies across five simulation benchmarks: RoboDojo, BEHAVIOR-1K, HumanoidArena, DexJoCo’s bimanual tasks, and LIBERO-Pro.
• One framework spanning single-arm, bimanual, dexterous, humanoid, and mobile robots, with new embodiments connected through APIs.
The craziest part: on each real-world embodiment, going from scratch to mastering the tasks took roughly two hours. No teleop. No model finetuning.
We’re open-sourcing the full framework AND every experiment session. We can’t wait to see what people build with it!
Project + sessions: robot-agent-exploration.com/
Paper: assets.robot-agent-explorati…
GitHub: N
Project fully led by @RunhanH and @xczhangxx who are absolutely awesome.
How it works, experiments, and robot videos below 🧵
(1/N)
Fu-En (Fred) Yang retweeted
ASPIRE is now open-source!
Since our initial release, we've made exciting progress:
🤖 Real-world robot manipulation with GPT-6 Astra.
🔄 Continual skill refinement on physical robots.
📈 Saturated simulation benchmarks: 99% on LIBERO-Pro (80 tasks), 100% on Robosuite (7 representative tasks), and 100% on BEHAVIOR-1K (2 representative tasks).
Code: github.com/NVlabs/ASPIRE
Project: research.nvidia.com/labs/gea…
🧵 More results in the thread below.
Fu-En (Fred) Yang retweeted
We scaled up language-action pre-training to 10k hours across robots and egocentric videos.
The key to facilitating motion transfer: using motion-level reasoning as a shared language across all the embodiments!
ego-lap.github.io/
What’s missing from robot foundation models? An explicit understanding of WHY motions work: why move in this direction, why place the object in this way, how hard to grasp the object. This motion-level reasoning bridges instructions to actions—and provides knowledge that can generalize across embodiments and scenes.
Pure behavior cloning supervises WHAT actions to take, but leaves models to infer WHY those actions work. This becomes especially limiting when learning from large-scale heterogeneous data, including egocentric human data, where humans and robots can take very different actions to achieve the same physical effect. We need models to extract the intent behind those motions, rather than simply recite them. Why leave that knowledge buried in trajectories when we can teach it explicitly?
Introducing EgoLAP: a shared language for human and robot motion. Language actions teach HOW to move; motion-level reasoning teaches WHY the motion works. Together, they unlock, to our knowledge, unprecedented gains in zero-shot performance for robot foundation models without additional robot demonstrations.
arxiv.org/abs/2610.08726
ego-lap.github.io/
Fu-En (Fred) Yang retweeted
Can we scale robot learning without scaling human labor? 📈
Demonstrations, resets, failure recovery, constant supervision—collecting robot data still takes substantial human effort.
🔑 Three keys let agents take on more of the work:
1. Real2Sim2Real: Agents explore and learn in digital twins, where trial and error is safer and cheaper.
2. Autonomous Recovery: Robots learn to recover from unexpected failures instead of waiting for human resets.
3. Human-on-Demand Learning: Multiple robots run in parallel, and humans step in only when autonomous recovery falls short.
🥑 Recova brings these together to get more robot learning out of every hour of human effort.
research.nvidia.com/labs/gea…
Joint work with @anjjei, Johan Bjorck, @ZhidingYu, @yin_hongxu, @jankautz, @DrJimFan, @yukez, and @Sifei30488L.
👇 Here's how it works 🧵
Fu-En (Fred) Yang retweeted
Our G1 robot found its way to the pantry, picked up a Pringles can, and handed it over. No pre-built map. No training for this task.
Behind it is a coding agent, writing, running, and fixing the robot's programs as it goes. And it self-evolves: what works is saved as code for the next task, no retraining needed. This is ASENA.
Joint work with @Isabella__Liu, @edmund_bu, Johan Bjorck, @yin_hongxu, @zhengyiluo, @jankautz, @DrJimFan, @yukez, and @Sifei30488L. 🧵👇
Fu-En (Fred) Yang retweeted
Can we build intelligent robots to assemble complex AI hardware? 🦾
At NVIDIA’s Seattle Robotics Lab, we’re tackling that question through GB300 tester-tray assembly, from fastening busbars to inserting delicate connectors. 👇
Fu-En (Fred) Yang retweeted
From simulation to real-world autonomy. 🤖
Dyna Robotics trained Dyna-2.1 in part with NVIDIA Isaac Lab, where thousands of simulated robots learned whole-body control in parallel. Learn how Dyna-2.1 handles an entire hotel laundry workflow from start to finish.
We are releasing Dyna-2.1, the first Physical Agent that achieves reliable super long-horizon whole-body autonomy. It combines our brand-new semi-humanoid hardware with an agentic system built around Dyna-2 to handle ultra-long real-world workflows.
Here is an uncut footage of Dyna-2.1 completing an entire hour-long laundry room workflow, just like a human does.
Fu-En (Fred) Yang retweeted
We are releasing Dyna-2.1, the first Physical Agent that achieves reliable super long-horizon whole-body autonomy. It combines our brand-new semi-humanoid hardware with an agentic system built around Dyna-2 to handle ultra-long real-world workflows.
Here is an uncut footage of Dyna-2.1 completing an entire hour-long laundry room workflow, just like a human does.