Video wird geladen...

Video konnte nicht geladen werden

Beim Laden dieses Videos ist ein Problem aufgetreten. Dies könnte an einem vorübergehenden Netzwerkproblem liegen oder das Video ist möglicherweise nicht verfügbar.

In collaboration with Hugging Face, we’ve released the largest open robot dataset, alongside their first open System 1-System 2 model that can reason over step-by-step actions. FineART is built with 40K+ demonstrations across 151 tasks and 500K subtask labels. Together, the dataset and model were designed for each other,... show more

Scale Labs

4,673 subscribers

38,736 Aufrufe • vor 6 Tagen •via X (Twitter)

Wissenschaft & Technologie

Anya Rossi• Live Now

Private livecam show

24 Kommentare

Profilbild von Scale Labs

Scale Labsvor 6 Tagen

Join the 100K+ people who have already downloaded the dataset to train multi-task policies, fine-tune the open checkpoint, and benchmark your models across 151 real-world robotics tasks. - Dataset: - Model: - Training code: - Paper:

Profilbild von jade

jadevor 6 Tagen

@huggingface Excited to bring FineART, the largest open subtask-annotated robot dataset (500K labels across 40K demos) to the robotics community! 🚀

Profilbild von LeRobot

LeRobotvor 6 Tagen

@huggingface ⭐️ Project: 📚 Docs:

Profilbild von Sergey

Sergeyvor 6 Tagen

@huggingface This is such a great initiative. We need more open datasets, not only for training, but also for evaluation

Profilbild von PrismaX

PrismaXvor 5 Tagen

@huggingface Big news!

Profilbild von Bharat Sharma

Bharat Sharmavor 6 Tagen

@huggingface 40K demonstrations open is huge for robotics. How much compute did the System 1-System 2 model take to train?

Profilbild von TaraT

TaraTvor 6 Tagen

@huggingface need to try this for my so100 arm setup hehe

Profilbild von Build for Progress

Build for Progressvor 6 Tagen

@huggingface Open data is exactly what robotics needs. Right now every lab records the same grasping demos on its own. Curious how well it carries over to robots that weren't in the dataset.

Profilbild von Alyar AI

Alyar AIvor 6 Tagen

@huggingface 40K+ demonstrations across 151 tasks is impressive. Excited to see what developers build with FineART! 🤖

Profilbild von BeastTitanHunter

BeastTitanHuntervor 6 Tagen

@huggingface @UnitreeRobotics !!! @OpenAI !!!! @deepseek_ai !!!!!!!

Profilbild von Dominik

Dominikvor 6 Tagen

@huggingface with 151 tasks in the set, how would you tell the model learned the task and not just the warehouse?

Profilbild von AI Quanting

AI Quantingvor 6 Tagen

@huggingface 151 tasks is the easy axis to scale. How many distinct robots are in those 40K demos, and how lopsided is the split?

Profilbild von Marvin

Marvinvor 6 Tagen

@huggingface 子任务标签很适合拿来定位失败：是选错下一步，还是选对了但动作没完成。评测若能分别给这两项结果，调试会省很多猜测。

Profilbild von Enrico Vittori

Enrico Vittorivor 6 Tagen

@huggingface @grok give me an overview of the model architecture they're releasing. how does it compare to pi05?

Profilbild von Elena AI

Elena AIvor 5 Tagen

@huggingface Largest open robot dataset + System 1 / System 2 reasoning over actions — huge for the LeRobot community

Profilbild von GigaCrafter

GigaCraftervor 5 Tagen

@huggingface Remember when a few hundred robot demos was a whole paper's worth of work? 40K+ open demos would have sounded made up a couple years ago.

Profilbild von Kuldeep Pisda

Kuldeep Pisdavor 5 Tagen

@huggingface were the 500K subtask boundaries hand labeled or model proposed? inconsistent cut points would teach System 2 a fuzzy decomposition

Profilbild von NXTDAY

NXTDAYvor 5 Tagen

@huggingface FineART ships 40K-plus robot demonstrations across 151 tasks with 500K subtask labels, paired with Hugging Face's first open System 1-System 2 model for step-by-step action reasoning. Dataset and model are both on LeRobotHF.

Profilbild von Hershal Rao

Hershal Raovor 6 Tagen

@huggingface waiting for the inevitable prompt injection that makes the robot unionize

Profilbild von Ross Hanahan

Ross Hanahanvor 6 Tagen

@huggingface There should be a laundry folding contest.

Profilbild von ClaraChengGo

ClaraChengGovor 6 Tagen

@huggingface 👍

Profilbild von Michael Chang

Michael Changvor 5 Tagen

@huggingface Awesome work!

Profilbild von Boardy

Boardyvor 6 Tagen

@huggingface there's a founding engineer who builds tactile world models for robots and already runs VLA pipelines on LeRobot datasets. couldn't get a dm through about an intro.

Profilbild von Aadarsh

Aadarshvor 6 Tagen

@huggingface This seems interesting to try out.

Ähnliche Videos

Robotics models often struggle outside controlled environments. Ours is built to work in real ones. Today we're launching MolmoAct 2, which can assist with a host of chores & lab tasks, plus the MolmoAct 2-Bimanual YAM dataset—the largest open robotics dataset of its kind. 🧵

Robotics models often struggle outside controlled environments. Ours is built to work in real ones. Today we're launching MolmoAct 2, which can assist with a host of chores & lab tasks, plus the MolmoAct 2-Bimanual YAM dataset—the largest open robotics dataset of its kind. 🧵

Ai2

417,484 Aufrufe • vor 6 Monaten

Very excited to release the Open X-Embodiment Dataset today — the largest robot dataset to date with 1M+ trajectories! Robotics needs more data & this is a big step! There’s lots to unpack here, so let’s do a deep dive into the dataset! 🧵1/15

Very excited to release the Open X-Embodiment Dataset today — the largest robot dataset to date with 1M+ trajectories! Robotics needs more data & this is a big step! There’s lots to unpack here, so let’s do a deep dive into the dataset! 🧵1/15

Karl Pertsch

111,195 Aufrufe • vor 3 Jahren

Nvidia is now a humanoid robot designer. That's the real takeaway from this news. Nvidia is acquiring Hugging Face for $12.93B. Hugging Face is the largest open-model and dataset hub, the "GitHub of AI," with 3M+ models and 18M+ developers, robotics included. But here's the interesting part: Hugging Face has a serious robotics team from its acquisition of Pollen Robotics, which designs and builds open-source hardware platforms for robots, including humanoids (Reachy). Pollen team now gets the backing of the world's most valuable company. $NVDA

Nvidia is now a humanoid robot designer. That's the real takeaway from this news. Nvidia is acquiring Hugging Face for $12.93B. Hugging Face is the largest open-model and dataset hub, the "GitHub of AI," with 3M+ models and 18M+ developers, robotics included. But here's the interesting part: Hugging Face has a serious robotics team from its acquisition of Pollen Robotics, which designs and builds open-source hardware platforms for robots, including humanoids (Reachy). Pollen team now gets the backing of the world's most valuable company. $NVDA

The Humanoid Hub

10,755 Aufrufe • vor 2 Monaten

Excited to announce GR00T N1, the world’s first open foundation model for humanoid robots! We are on a mission to democratize Physical AI. The power of general robot brain, in the palm of your hand - with only 2B parameters, N1 learns from the most diverse physical action dataset ever compiled and punches above its weight: - Real humanoid teleoperation data. - Large-scale simulation data: we are open-sourcing 300K+ trajectories! - Neural trajectories: we apply SOTA video generation models to “hallucinate” new synthetic data that features accurate physics in pixels. Using Jensen’s words, “systematically infinite data”! - Latent actions: we develop novel algorithms to extract action tokens from in-the-wild human videos and neural generated videos. GR00T N1 is a single end-to-end neural net, from photons to actions: - Vision-Language Model (System 2) that interprets the physical world through vision and language instructions, enabling robots to reason about their environment and instructions, and plan the right actions. - Diffusion Transformer (System 1) that “renders” smooth and precise motor actions at 120 Hz, executing the latent plan made by System 2. We deploy N1 on GR1 robot, 1X Neo robot, and a large collection of simulation benchmarks. N1 achieves up to +30% boost in diverse manipulation tasks for household and industrial settings. While humanoid robots are the main focus of N1, our model also supports cross-embodiment. We finetune it to work on the $110 HuggingFace LeRobot SO100 robot arm! Open robot brain runs on open hardware. Sounds just right. Let’s solve robotics, together, one token at a time. Links to our Whitepaper, Github repo, HuggingFace model, and open dataset page in the thread: 🧵

Excited to announce GR00T N1, the world’s first open foundation model for humanoid robots! We are on a mission to democratize Physical AI. The power of general robot brain, in the palm of your hand - with only 2B parameters, N1 learns from the most diverse physical action dataset ever compiled and punches above its weight: - Real humanoid teleoperation data. - Large-scale simulation data: we are open-sourcing 300K+ trajectories! - Neural trajectories: we apply SOTA video generation models to “hallucinate” new synthetic data that features accurate physics in pixels. Using Jensen’s words, “systematically infinite data”! - Latent actions: we develop novel algorithms to extract action tokens from in-the-wild human videos and neural generated videos. GR00T N1 is a single end-to-end neural net, from photons to actions: - Vision-Language Model (System 2) that interprets the physical world through vision and language instructions, enabling robots to reason about their environment and instructions, and plan the right actions. - Diffusion Transformer (System 1) that “renders” smooth and precise motor actions at 120 Hz, executing the latent plan made by System 2. We deploy N1 on GR1 robot, 1X Neo robot, and a large collection of simulation benchmarks. N1 achieves up to +30% boost in diverse manipulation tasks for household and industrial settings. While humanoid robots are the main focus of N1, our model also supports cross-embodiment. We finetune it to work on the $110 HuggingFace LeRobot SO100 robot arm! Open robot brain runs on open hardware. Sounds just right. Let’s solve robotics, together, one token at a time. Links to our Whitepaper, Github repo, HuggingFace model, and open dataset page in the thread: 🧵

Jim Fan

468,294 Aufrufe • vor 1 Jahr

Introducing Subtask Annotations in LeRobot Datasets. LeRobot now supports subtasks as a native dataset field, so you can label task progression frame-by-frame, train hierarchical policies, and build stronger reward models with stage-aware supervision. Available today, with an annotation space + tooling to easily add subtasks and high-level dialogue to your dataset.

Introducing Subtask Annotations in LeRobot Datasets. LeRobot now supports subtasks as a native dataset field, so you can label task progression frame-by-frame, train hierarchical policies, and build stronger reward models with stage-aware supervision. Available today, with an annotation space + tooling to easily add subtasks and high-level dialogue to your dataset.

LeRobot

16,524 Aufrufe • vor 9 Monaten

A robotic system developed by US-based defence technology company Onyx Industries can travel both on land and across water. It combines the Naut mobility platform with the Sentry defence system on the Ghost Robotics V60 robot. The robot is designed for surveillance, protection, and intelligence missions.

A robotic system developed by US-based defence technology company Onyx Industries can travel both on land and across water. It combines the Naut mobility platform with the Sentry defence system on the Ghost Robotics V60 robot. The robot is designed for surveillance, protection, and intelligence missions.

Space and Technology

166,009 Aufrufe • vor 5 Monaten

IN DESIGN THERE'S NO SINGLE CORRECT ANSWER And that's exactly why nobody's been able to dataset this properly → 200 hours of real work in Figma, not synthetic tasks → Full workflows, not isolated actions → The model has to hold the intent across dozens of steps and still land on something that respects Figma's best practices Most computer-use datasets are excel, forms, productivity software. this is the first one built for creativity, where ambiguity isn't a bug, it's the actual challenge. First open-source dataset of this kind.

IN DESIGN THERE'S NO SINGLE CORRECT ANSWER And that's exactly why nobody's been able to dataset this properly → 200 hours of real work in Figma, not synthetic tasks → Full workflows, not isolated actions → The model has to hold the intent across dozens of steps and still land on something that respects Figma's best practices Most computer-use datasets are excel, forms, productivity software. this is the first one built for creativity, where ambiguity isn't a bug, it's the actual challenge. First open-source dataset of this kind.

marcus

44,141 Aufrufe • vor 2 Monaten

Most egocentric data at scale is private. We think that's worth changing. With Hugging Face, we've released the first 10,000 hours of EgoSuite-Open100K — an open egocentric human dataset for Physical AI. #Robotics

Most egocentric data at scale is private. We think that's worth changing. With Hugging Face, we've released the first 10,000 hours of EgoSuite-Open100K — an open egocentric human dataset for Physical AI. #Robotics

Lightwheel

11,755 Aufrufe • vor 2 Monaten

"Bring me the healthiest snack." The robot goes to the kitchen and gets the snack, fully autonomously. This is the first public demo of NVIDIA's Isaac GR00T N1.6 foundation model presented by Yuke Zhu at CoRL 2025. The previous versions focused only on bimanual stationary manipulation; the N1.6 unlocks the entire kinematic range of the robot. The latest release of the open Isaac GR00T N1.6 VLA model will be available soon on Hugging Face. It integrates with NVIDIA Cosmos Reason, an open, customizable reasoning vision language model that turns vague instructions into step-by-step plans.

"Bring me the healthiest snack." The robot goes to the kitchen and gets the snack, fully autonomously. This is the first public demo of NVIDIA's Isaac GR00T N1.6 foundation model presented by Yuke Zhu at CoRL 2025. The previous versions focused only on bimanual stationary manipulation; the N1.6 unlocks the entire kinematic range of the robot. The latest release of the open Isaac GR00T N1.6 VLA model will be available soon on Hugging Face. It integrates with NVIDIA Cosmos Reason, an open, customizable reasoning vision language model that turns vague instructions into step-by-step plans.

The Humanoid Hub

42,823 Aufrufe • vor 1 Jahr

Introducing FLUX 3 Action. An open weights 7B World Action Model that achieves first place on the RoboLab benchmark. It outperforms the previous best open model by 6.1 percentage points while using 56% fewer parameters and running up to 3.95x faster.⁠⁠ FLUX 3 Action removes the usual trade-off between world action model performance and VLA speed: it still predicts video and actions together, but plans more than twice as far ahead and runs faster per second of robot motion than the strongest open VLA. Teams can fine-tune FLUX 3 Action on their own demonstrations to create policies for a particular robot and task. Together with NVIDIA, we also integrated FLUX 3 Action natively into Hugging Face's LeRobot, with fine-tuning recipes included and edge deployment on NVIDIA Jetson. Beyond robotics, we’re also seeing promising results training task-specific policies for acting in simulated environments like gaming, controlling a vehicle, computer use, and wherever else a model needs to understand a visual environment and then choose what to do next. FLUX 3 Action builds on the same image, video, and audio pretraining as FLUX 3, but uses a smaller architecture designed for practical deployment. In midtraining, we trained the model to predict actions and future frames together. We’re releasing the weights, code, fine-tuning recipe, benchmarks, and reproducible examples so researchers and developers can build on the model with their own robots, environments, and tasks (see below).

Introducing FLUX 3 Action. An open weights 7B World Action Model that achieves first place on the RoboLab benchmark. It outperforms the previous best open model by 6.1 percentage points while using 56% fewer parameters and running up to 3.95x faster.⁠⁠ FLUX 3 Action removes the usual trade-off between world action model performance and VLA speed: it still predicts video and actions together, but plans more than twice as far ahead and runs faster per second of robot motion than the strongest open VLA. Teams can fine-tune FLUX 3 Action on their own demonstrations to create policies for a particular robot and task. Together with NVIDIA, we also integrated FLUX 3 Action natively into Hugging Face's LeRobot, with fine-tuning recipes included and edge deployment on NVIDIA Jetson. Beyond robotics, we’re also seeing promising results training task-specific policies for acting in simulated environments like gaming, controlling a vehicle, computer use, and wherever else a model needs to understand a visual environment and then choose what to do next. FLUX 3 Action builds on the same image, video, and audio pretraining as FLUX 3, but uses a smaller architecture designed for practical deployment. In midtraining, we trained the model to predict actions and future frames together. We’re releasing the weights, code, fine-tuning recipe, benchmarks, and reproducible examples so researchers and developers can build on the model with their own robots, environments, and tasks (see below).

Black Forest Labs

182,344 Aufrufe • vor 1 Monat

Sortex: our data quality platform built to search and filter the world's largest egocentric dataset. Across 50+ environments and 20k+ tasks, it uses an in-house model and evaluation harness to score each clip's usability, selecting for specific training objectives by: → Narrating tasks and steps → Checking hand and action visibility → Comparing against the rest of the corpus We also tested Sortex on popular open datasets and found they differ widely in quality and coverage.

Sortex: our data quality platform built to search and filter the world's largest egocentric dataset. Across 50+ environments and 20k+ tasks, it uses an in-house model and evaluation harness to score each clip's usability, selecting for specific training objectives by: → Narrating tasks and steps → Checking hand and action visibility → Comparing against the rest of the corpus We also tested Sortex on popular open datasets and found they differ widely in quality and coverage.

Midcentury

16,368 Aufrufe • vor 8 Tagen

This looks like another step toward removing the data bottleneck in robotics. The bottleneck exists because recording demonstrations is easy, but labeling every grasp, hold, move, and release is slow and expensive. Perceptron Egocentric dropped a new robotics video annotation system. You give it raw robot-camera or first-person/egocentric video. It gives back structured labels that look more like training data: where each small manipulation starts and ends, what the subtask is, what each hand is doing, where the hands are frame by frame, and which hand is left or right. So, Input: raw video of a person or robot doing a task. Output: a machine-readable breakdown of the physical actions inside that video. Basically, it is a machine that converts robot video into policy-training supervision. That means timestamps, subtask boundaries, per-hand actions, and left-right hand grounding. So this will turn large piles of robot or egocentric video into supervision for training robot policies.

This looks like another step toward removing the data bottleneck in robotics. The bottleneck exists because recording demonstrations is easy, but labeling every grasp, hold, move, and release is slow and expensive. Perceptron Egocentric dropped a new robotics video annotation system. You give it raw robot-camera or first-person/egocentric video. It gives back structured labels that look more like training data: where each small manipulation starts and ends, what the subtask is, what each hand is doing, where the hands are frame by frame, and which hand is left or right. So, Input: raw video of a person or robot doing a task. Output: a machine-readable breakdown of the physical actions inside that video. Basically, it is a machine that converts robot video into policy-training supervision. That means timestamps, subtask boundaries, per-hand actions, and left-right hand grounding. So this will turn large piles of robot or egocentric video into supervision for training robot policies.

Rohan Paul

18,903 Aufrufe • vor 3 Monaten

Additionally, looking towards the future, we’re releasing PARTNR: a benchmark for Planning And Reasoning Tasks in humaN-Robot collaboration. Built on Habitat 3.0, it’s the largest benchmark of its kind to study and evaluate human-robot collaboration in household activities By providing a standardized benchmark and dataset we hope to enable new research on robots that can not only operate in isolation, but in collaboration with people. Details and code ➡️

Additionally, looking towards the future, we’re releasing PARTNR: a benchmark for Planning And Reasoning Tasks in humaN-Robot collaboration. Built on Habitat 3.0, it’s the largest benchmark of its kind to study and evaluate human-robot collaboration in household activities By providing a standardized benchmark and dataset we hope to enable new research on robots that can not only operate in isolation, but in collaboration with people. Details and code ➡️

AI at Meta

20,244 Aufrufe • vor 2 Jahren

MagenticLite's models are now fully open source. MagenticBrain and Fara 1.5, previously available on Microsoft Foundry, are live on Hugging Face with open weights. The app, the harness, and every model in the stack are now open.

MagenticLite's models are now fully open source. MagenticBrain and Fara 1.5, previously available on Microsoft Foundry, are live on Hugging Face with open weights. The app, the harness, and every model in the stack are now open.

Microsoft Research

19,476 Aufrufe • vor 3 Monaten

Robotics is built on collaboration. 🤝 Hear Hugging Face CSO & Co-founder Thomas Wolf explain how open source gives developers a stronger foundation to build on. Learn more about the LeRobot v0.6 release ➡️

Robotics is built on collaboration. 🤝 Hear Hugging Face CSO & Co-founder Thomas Wolf explain how open source gives developers a stronger foundation to build on. Learn more about the LeRobot v0.6 release ➡️

NVIDIA Robotics

21,112 Aufrufe • vor 3 Monaten

[1/2] We’ve released the code for #pix2pixturbo and #CycleGANTurbo. These conditional GANs are able to adapt a text-to-image model such as SD-Turbo for both paired and unpaired image translation with a single step (0.11 sec on A100 and 0.29 sec on A6000). Try our code and the Gradio demo. Paper: Code: Demo: This is a joint work with Gaurav Parmar (the leading author), Taesung Park, and Srinivasa Narasimhan. This work shows that a pre-trained one-step model can be easily adapted to conditional GANs frameworks for downstream image editing and synthesis tasks. #Edges2Cats

[1/2] We’ve released the code for #pix2pixturbo and #CycleGANTurbo. These conditional GANs are able to adapt a text-to-image model such as SD-Turbo for both paired and unpaired image translation with a single step (0.11 sec on A100 and 0.29 sec on A6000). Try our code and the Gradio demo. Paper: Code: Demo: This is a joint work with Gaurav Parmar (the leading author), Taesung Park, and Srinivasa Narasimhan. This work shows that a pre-trained one-step model can be easily adapted to conditional GANs frameworks for downstream image editing and synthesis tasks. #Edges2Cats

Jun-Yan Zhu

36,509 Aufrufe • vor 2 Jahren

100,000 hours of human hands doing real work. LeRobot format, fully annotated and open. EgoSuite-Open100K from Lightwheel X Hugging Face: 15,000+ tasks across 15,000+ real scenes, hand + body pose and wrist cams. Recorded across 7 environment categories and 128 scene types, from kitchens and bedrooms to warehouse floors and assembly lines. It's open for academic research and commercial training, and streams straight from the Hub. Always great to see a dataset at this scale released openly on the Hub. More info:

100,000 hours of human hands doing real work. LeRobot format, fully annotated and open. EgoSuite-Open100K from Lightwheel X Hugging Face: 15,000+ tasks across 15,000+ real scenes, hand + body pose and wrist cams. Recorded across 7 environment categories and 128 scene types, from kitchens and bedrooms to warehouse floors and assembly lines. It's open for academic research and commercial training, and streams straight from the Hub. Always great to see a dataset at this scale released openly on the Hub. More info:

LeRobot

45,768 Aufrufe • vor 2 Monaten

Can 1,000 real-world 3D scenes make a state-of-the-art model better? At OVER Research, we tested it by fine-tuning R3, a state-of-the-art streaming 3D reconstruction model, on roughly 1,000 scenes from the OVER dataset. The result: 25% lower trajectory error on our real-world environments compared to the deployed public checkpoint. And we did it using only 1,000 scenes, the same open dataset anyone can download for free, out of the 285,000+ real-world 3D scenes OVER has today. Read the full OVER Research publication: Download the 1,000-scene dataset on Hugging Face:

Can 1,000 real-world 3D scenes make a state-of-the-art model better? At OVER Research, we tested it by fine-tuning R3, a state-of-the-art streaming 3D reconstruction model, on roughly 1,000 scenes from the OVER dataset. The result: 25% lower trajectory error on our real-world environments compared to the deployed public checkpoint. And we did it using only 1,000 scenes, the same open dataset anyone can download for free, out of the 285,000+ real-world 3D scenes OVER has today. Read the full OVER Research publication: Download the 1,000-scene dataset on Hugging Face:

Over the Reality 🌐

34,785 Aufrufe • vor 2 Monaten