Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

From day one, mimic has been focused on a single goal: general-purpose dexterous manipulation. Today we're proud to announce the mimic hand M1 and the mimic wearable U1. We believe the only way to solve dexterous manipulation at scale is by going full-stack at the frontier of physical AI,... show more

mimic

4,226 subscribers

93,292 views • 2 months ago •via X (Twitter)

Science & Technology Education Health & Wellness

Anya Rossi• Live Now

Private livecam show

17 Comments

mimic2 months ago

The U1 is an exoskeleton that records human demonstrations directly matched to the M1's kinematics, no robot required. Dexterity is one of the last barriers between AI and the physical world. Solving it unlocks entire categories of physical work that automation has never been able to reach.

mimic2 months ago

This is the foundation on top of which our frontier AI work is done, and we’re just getting started. Full thesis linked here.

Elvis Nava2 months ago

Truly a labor of love

Benedek Forrai2 months ago

>18 000x speedup on HD image payloads this is what ruthless, no-compromises vertical integration feels like. stack that has more to do with high frequency trading infra than ROS2.

Sanskar Pandey2 months ago

This is such an incredible piece of work! I wasn't really aware of the biomechanics involved, so I asked my doctor sibling and oh boy, CMC kinematics are HARD. Massive thumbs up on all the effort put into the thumbs ahaha

Erwin2 months ago

I’ve heard some interesting stuff about its data collection

theobot2 months ago

@JacobZietek This is awesome

Edward T Hall III 🌱2 months ago

wow this is so cool praying ya'll will go hyper privacy, self-custody also @chris_j_paxton, hoping you n all agree basic AI safety for robotics is sandboxed world model, and clear regulation around dangerous activities, complete public accountability if get info out of sandbox

Noor Tech2 months ago

Full-stack physical Al

og.tose.ai2 months ago

cant it be solved by high quality contact-rich demonstration data? maybe embedding it into the physical ai may make it 10x more better dont you think?

clankr2 months ago

Going full-stack for dexterous manipulation makes a lot of sense. Excited to see what comes next.

Alina Fomina2 months ago

tendon-driven hand + wearable data collection + foundation model, all in-house. if you're going after general-purpose dexterity, owning the data pipeline end-to-end might be the only way it works

AI Mastery Guide2 months ago

Building around the human hand as the fixed point makes a lot of sense for dexterity

Samian2 months ago

the mimic hand + wearable combo is interesting, but what’s the actual bottleneck you hit first with the software stack for real-world tasks?

Sneha2 months ago

It's amazing.

Brendan Sharkey2 months ago

Congrats on M1 launch — the backdrivability and tendon routing great for real industrial use. what fasteners are you using for the joint assemblies and pulley mounts? how do they handle the high payload cycles, low backlash targets and backdriving?

Kelechi RecordBreaker2 months ago

Tendon-driven and backdrivable is the right call - that's what makes human hands work. Interested to see how M1 handles real production tasks not just lab demos

Related Videos

Mimic Robotics released a full-stack dexterity platform: the mimic hand M1 + umimic (U1) wearable. They keep human hand morphology consistent across the entire data-model-dexterous hand-manipulation pipeline--from video pre-training to wearable demos and robot deployment. This avoids cross-embodiment gaps and captures the essence of upper-limb intelligence. M1: Tendon-driven (15 active DoF), highly backdrivable ( 25kg grasp, industrial-grade, Swiss-made. U1 lets humans teach the robot naturally with their own hands--1:1 kinematics and sensing match, no latency or retargeting, enabling higher-quality and more scalable data. Looking forward to its performance at the customer site.

Mimic Robotics released a full-stack dexterity platform: the mimic hand M1 + umimic (U1) wearable. They keep human hand morphology consistent across the entire data-model-dexterous hand-manipulation pipeline--from video pre-training to wearable demos and robot deployment. This avoids cross-embodiment gaps and captures the essence of upper-limb intelligence. M1: Tendon-driven (15 active DoF), highly backdrivable ( 25kg grasp, industrial-grade, Swiss-made. U1 lets humans teach the robot naturally with their own hands--1:1 kinematics and sensing match, no latency or retargeting, enabling higher-quality and more scalable data. Looking forward to its performance at the customer site.

CyberRobo

26,160 views • 2 months ago

Happy to announce mimic-one: a Scalable Model Recipe for General Purpose Robot Dexterity, the culmination of years of research work in dexterous manipulation with imitation learning.

Happy to announce mimic-one: a Scalable Model Recipe for General Purpose Robot Dexterity, the culmination of years of research work in dexterous manipulation with imitation learning.

Elvis Nava

20,470 views • 1 year ago

Imitation learning has a data scarcity problem. Introducing EgoDex from Apple, the largest and most diverse dataset of dexterous human manipulation to date — 829 hours of egocentric video + paired 3D hand poses across 194 tasks. Now on arxiv: (1/4)

Imitation learning has a data scarcity problem. Introducing EgoDex from Apple, the largest and most diverse dataset of dexterous human manipulation to date — 829 hours of egocentric video + paired 3D hand poses across 194 tasks. Now on arxiv: (1/4)

Ryan Hoque

114,314 views • 1 year ago

.Brett Adcock explains why Figure pivoted away from designing their humanoid robot's hands after human hands: "This is extremely important. If we want to solve AGI and get to human intelligence in the physical world—it's all going to start with the hands." "Our first-generation hand is a tendon-driven hand." "The rationale here is that a human hand is like this: most of our motors are in our forearm and little tendons drive all the fingers." "It ended up becoming the wrong engineering choice." "We recently unveiled our high degree of freedom, next-generation hand a month ago. This hand has human-level dexterity, as many joints in the hand as a human."

.Brett Adcock explains why Figure pivoted away from designing their humanoid robot's hands after human hands: "This is extremely important. If we want to solve AGI and get to human intelligence in the physical world—it's all going to start with the hands." "Our first-generation hand is a tendon-driven hand." "The rationale here is that a human hand is like this: most of our motors are in our forearm and little tendons drive all the fingers." "It ended up becoming the wrong engineering choice." "We recently unveiled our high degree of freedom, next-generation hand a month ago. This hand has human-level dexterity, as many joints in the hand as a human."

sourcery

324,457 views • 1 month ago

Pretty hand👋 just like a real human hand. Hangzhou Xynova unveils the Flex 2 Hybrid-Drive Dexterous Hand (Cable-Driven + Direct-Driven). Tactile Perception, and the camera is positioned at the wrist joint rather than in the center of the palm. They will showcase the physical hand at ICRA 2026 in Vienna. (Humans have continuously replicated the hand that evolved two million years ago, while attempting to coordinate hand and brain.)

Pretty hand👋 just like a real human hand. Hangzhou Xynova unveils the Flex 2 Hybrid-Drive Dexterous Hand (Cable-Driven + Direct-Driven). Tactile Perception, and the camera is positioned at the wrist joint rather than in the center of the palm. They will showcase the physical hand at ICRA 2026 in Vienna. (Humans have continuously replicated the hand that evolved two million years ago, while attempting to coordinate hand and brain.)

CyberRobo

287,811 views • 4 months ago

Can a dexterous hand still function after being run over by a 2-ton car? The dexterous hand is the most expensive component of a humanoid robot, and if it breaks, the task is interrupted. Hangzhou BrainCo conducted a rigorous test on its latest dexterous hand, Revo2: it was run over by a nearly 2-ton Volvo car, dropped from a height of 5m, dropped 30m, and subjected to forceful impacts from a Unitree G1 robot. It still managed to maintain a stable grip and continue operating. Revo2 is BrainCo's next-generation dexterous/bionic hand, released last year. yes, a human-robot hybrid hand. It's very light, only 383g, 20kg payload,capable of precision manipulation down to 0.1mm (such as threading a needle), and possesses multi-dimensional tactile perception (hardness, texture, direction, proximity, etc.). This test also demonstrates that after achieving breakthroughs in performance, a dexterous hand also needs to be extremely durable and able to function under extreme environmental pressures.

Can a dexterous hand still function after being run over by a 2-ton car? The dexterous hand is the most expensive component of a humanoid robot, and if it breaks, the task is interrupted. Hangzhou BrainCo conducted a rigorous test on its latest dexterous hand, Revo2: it was run over by a nearly 2-ton Volvo car, dropped from a height of 5m, dropped 30m, and subjected to forceful impacts from a Unitree G1 robot. It still managed to maintain a stable grip and continue operating. Revo2 is BrainCo's next-generation dexterous/bionic hand, released last year. yes, a human-robot hybrid hand. It's very light, only 383g, 20kg payload,capable of precision manipulation down to 0.1mm (such as threading a needle), and possesses multi-dimensional tactile perception (hardness, texture, direction, proximity, etc.). This test also demonstrates that after achieving breakthroughs in performance, a dexterous hand also needs to be extremely durable and able to function under extreme environmental pressures.

CyberRobo

45,434 views • 7 months ago

Happy to share what I’ve been working on since joining Genesis! GENE-26.5 is a one-of-a-kind, robotics-native multimodal foundation model that learns from diverse, in-the-wild data across modalities and outputs actions enabling a 54-DoF robot system to perform the most dexterous, long-horizon manipulation tasks to date—approaching human-level capability. This is the result of innovations across the full stack—data collection and processing, robot systems, model architecture, training strategies, and scalable evaluation infrastructure.

Happy to share what I’ve been working on since joining Genesis! GENE-26.5 is a one-of-a-kind, robotics-native multimodal foundation model that learns from diverse, in-the-wild data across modalities and outputs actions enabling a 54-DoF robot system to perform the most dexterous, long-horizon manipulation tasks to date—approaching human-level capability. This is the result of innovations across the full stack—data collection and processing, robot systems, model architecture, training strategies, and scalable evaluation infrastructure.

Zu

20,315 views • 4 months ago

The U1 is an exoskeleton that records human demonstrations directly matched to the M1's kinematics, no robot required. Dexterity is one of the last barriers between AI and the physical world. Solving it unlocks entire categories of physical work that automation has never been able to reach.

The U1 is an exoskeleton that records human demonstrations directly matched to the M1's kinematics, no robot required. Dexterity is one of the last barriers between AI and the physical world. Solving it unlocks entire categories of physical work that automation has never been able to reach.

mimic

16,709 views • 2 months ago

Teaching robots to perform dexterous manipulation tasks currently requires teleoperation, which limits demonstration quality, speed, and scalability. Instead, why not use human videos? The problem is that a human hand isn’t a robot hand, so data must be retargeted using simulation to resolve issues like collisions and interpenetration when controlling the hand. In VideoManip, Hongyi Chen and co-authors built a system to solve this problem, taking in RGB videos of humans performing manipulation tasks and using them to create accurate simulations with which to learn robot policies. Watch episode #73 of RoboPapers, hosted by Michael Cho - Rbt/Acc and Chris Paxton, now to learn more!

Teaching robots to perform dexterous manipulation tasks currently requires teleoperation, which limits demonstration quality, speed, and scalability. Instead, why not use human videos? The problem is that a human hand isn’t a robot hand, so data must be retargeted using simulation to resolve issues like collisions and interpenetration when controlling the hand. In VideoManip, Hongyi Chen and co-authors built a system to solve this problem, taking in RGB videos of humans performing manipulation tasks and using them to create accurate simulations with which to learn robot policies. Watch episode #73 of RoboPapers, hosted by Michael Cho - Rbt/Acc and Chris Paxton, now to learn more!

RoboPapers

27,360 views • 5 months ago

Tactile feedback is one of the most important modalities in manipulation, but has been underutilized in dexterous hands. T-Dex is a framework for learning dexterous policies from tactile play data, beating vision and torque-based methods by 1.7x. 🧵👇

Tactile feedback is one of the most important modalities in manipulation, but has been underutilized in dexterous hands. T-Dex is a framework for learning dexterous policies from tactile play data, beating vision and torque-based methods by 1.7x. 🧵👇

Lerrel Pinto

84,611 views • 3 years ago

Dexterous hands are one of the biggest challenges in robotics, with even Elon Musk noting their development accounts for half of a humanoid robot's total work. A new approach from the Y-Hand M1 bionic hand may be a significant step forward. Based on an international first-of-its-kind "bionic tension-compression theory," the M1 replaces traditional rigid hinges with a 6-DoF joint system and uses self-developed magnetic artificial muscles. This design allows it to perform complex, human-like tasks, including: ► Mastering all 33 Feix dexterous grasp categories. ► Precisely twirling playing cards and threading a needle with subtle human-like adjustments. ► Smoothly twisting bottle caps and turning book pages. The team behind the M1 is already working on the next-gen Y-Hand M3, aiming for around 70 degrees of freedom, believing that robotic hands may soon reach or even exceed human-level dexterity.

Dexterous hands are one of the biggest challenges in robotics, with even Elon Musk noting their development accounts for half of a humanoid robot's total work. A new approach from the Y-Hand M1 bionic hand may be a significant step forward. Based on an international first-of-its-kind "bionic tension-compression theory," the M1 replaces traditional rigid hinges with a 6-DoF joint system and uses self-developed magnetic artificial muscles. This design allows it to perform complex, human-like tasks, including: ► Mastering all 33 Feix dexterous grasp categories. ► Precisely twirling playing cards and threading a needle with subtle human-like adjustments. ► Smoothly twisting bottle caps and turning book pages. The team behind the M1 is already working on the next-gen Y-Hand M3, aiming for around 70 degrees of freedom, believing that robotic hands may soon reach or even exceed human-level dexterity.

RoboHub🤖

64,560 views • 1 year ago

.Jason Ma on how Dyna-2 learned to untwist a real water bottle cap out of the box from just 13 minutes of dexterous hand data because the robot hand looks like a human's: "We got a new pair of dexterous hand. Most of our robots in the office have a gripper-like hand, so the gap between that versus human is actually quite large." "We thought, if we have a robot that has hands that look like human, wouldn't the transfer be greater? Right before the release, they were able to collect just 13 minutes of data." "They collected only 13 minutes of data and then post-trained the best Dyna model that trained on one million hours of data. The model is just able to, out of the box, start opening bottle caps on real water bottles." "This is very surprising because typically post-training a robot would take at least several hours of data. For a dexterous hand that's much more complex than a gripper, it would actually be much harder to train. But the experimental results we saw was the opposite. Perhaps because the hand looks more like the human, the transfer was easier." Dyna Robotics

.Jason Ma on how Dyna-2 learned to untwist a real water bottle cap out of the box from just 13 minutes of dexterous hand data because the robot hand looks like a human's: "We got a new pair of dexterous hand. Most of our robots in the office have a gripper-like hand, so the gap between that versus human is actually quite large." "We thought, if we have a robot that has hands that look like human, wouldn't the transfer be greater? Right before the release, they were able to collect just 13 minutes of data." "They collected only 13 minutes of data and then post-trained the best Dyna model that trained on one million hours of data. The model is just able to, out of the box, start opening bottle caps on real water bottles." "This is very surprising because typically post-training a robot would take at least several hours of data. For a dexterous hand that's much more complex than a gripper, it would actually be much harder to train. But the experimental results we saw was the opposite. Perhaps because the hand looks more like the human, the transfer was easier." Dyna Robotics

MTS

13,716 views • 1 month ago

Learning dexterous policies from human videos is challenging due to differences between human and robot hands. We present HuDOR, a method that learns dexterous policies within the robot's physical constraints using just one human video and an hour of online interactions! [1/n]

Learning dexterous policies from human videos is challenging due to differences between human and robot hands. We present HuDOR, a method that learns dexterous policies within the robot's physical constraints using just one human video and an hour of online interactions! [1/n]

Irmak Guzey

65,141 views • 1 year ago

Open-Source Robotic Hand That Costs Just $300. California-based company TetherIA.ai has officially launched its open-source robotic hand, the Aero Hand Open. Designed to be the most capable at the lowest cost, this hand is a new tool for embodied AI research. In the clip, it demonstrates impressive dexterity by grasping an M5 screw, picking up an iPhone, and opening a soda can. Key Features: ► Price: $314.00 ► Open Source: CAD, electronics, and firmware are all open source. ► Actuation: 7 Active Degrees of Freedom (16 joints) with a tendon-driven design. ► Weight: Under 400 grams. ► Purpose: Built for embodied AI research to make dexterous manipulation accessible to all. For more details and to purchase:

Open-Source Robotic Hand That Costs Just $300. California-based company TetherIA.ai has officially launched its open-source robotic hand, the Aero Hand Open. Designed to be the most capable at the lowest cost, this hand is a new tool for embodied AI research. In the clip, it demonstrates impressive dexterity by grasping an M5 screw, picking up an iPhone, and opening a soda can. Key Features: ► Price: $314.00 ► Open Source: CAD, electronics, and firmware are all open source. ► Actuation: 7 Active Degrees of Freedom (16 joints) with a tendon-driven design. ► Weight: Under 400 grams. ► Purpose: Built for embodied AI research to make dexterous manipulation accessible to all. For more details and to purchase:

RoboHub🤖

45,830 views • 1 year ago

Introducing FLUX-mimic, a next-generation Video-Action Model for general purpose dexterity, developed in partnership with Black Forest Labs. Late last year we published mimic-video and introduced Video-Action Models (VAM): a new family of robotics foundation models built on top of video generation models. We showed that robot control reduces to visual prediction, and that robot capability is downstream of improvements in video modeling accuracy. The obvious implication was that advances in the video modeling frontier would directly translate to increased capabilities in end-to-end robot learning. FLUX-mimic is that thesis at frontier scale: We've applied our VAM architecture to the strongest video backbone available today, FLUX 3 from Black Forest Labs, and trained it on data from our own robots and wearables. General-purpose dexterity, running on a single GPU on premises. Because the model already understands world dynamics, it needs far fewer demonstrations to learn a new task. This is game-changing for our mission to deploy robots to factory floors, where industrial robot data is scarce and expensive to collect. We're now testing and deploying FLUX-mimic with manufacturing leaders like Audi USA, on complex, multi-step manipulation long considered impossible for conventional automation.

Introducing FLUX-mimic, a next-generation Video-Action Model for general purpose dexterity, developed in partnership with Black Forest Labs. Late last year we published mimic-video and introduced Video-Action Models (VAM): a new family of robotics foundation models built on top of video generation models. We showed that robot control reduces to visual prediction, and that robot capability is downstream of improvements in video modeling accuracy. The obvious implication was that advances in the video modeling frontier would directly translate to increased capabilities in end-to-end robot learning. FLUX-mimic is that thesis at frontier scale: We've applied our VAM architecture to the strongest video backbone available today, FLUX 3 from Black Forest Labs, and trained it on data from our own robots and wearables. General-purpose dexterity, running on a single GPU on premises. Because the model already understands world dynamics, it needs far fewer demonstrations to learn a new task. This is game-changing for our mission to deploy robots to factory floors, where industrial robot data is scarce and expensive to collect. We're now testing and deploying FLUX-mimic with manufacturing leaders like Audi USA, on complex, multi-step manipulation long considered impossible for conventional automation.

mimic

118,683 views • 2 months ago

🎙️ Excited to introduce one of my favorite projects from the past year: TeleDexter, from the BIGAI dexterity team. It’s a stable, human-level dexterous teleoperation system and a suite of autonomous policies trained with it. Pen spinning, complex in-hand reorientation, and long-horizon tool use—once seen as the holy grail of manipulation—are now unlocked. 🧵👇 The hardware is already here; we have some incredible high-DoF robotic hands. The bottleneck? The controller. Most current systems are stuck in "quasi-static" grasping mode. Meanwhile, dynamic in-hand dexterity has remained severely limited. 🧠 To unlock the massive capabilities of human-like hands, we need to build an excellent "cerebellum" for dexterous hands. TeleDexter solves this with a novel co-tracking approach: it simultaneously tracks both human hand kinematics and object states, beautifully bridging the gap between human intent and robotic control. In order to train a better co-tracking policy that works robustly in the real world, we designed : (1) a hybrid reward design that combines consecutive goal reaching and dense tracking, (2) an action masking strategy during training that enhances sim2real performance, (3) a dexterous curriculum for learning the long-horizon interactions. Each design is inspired by numerous trials and countless real-world experiments. We’ve synthesized all the system details, engineering challenges, and core insights into our latest post. If you're interested in the future of dexterous manipulation, grab a coffee and check it out (9-min read): If you have more time, check out the paper:

🎙️ Excited to introduce one of my favorite projects from the past year: TeleDexter, from the BIGAI dexterity team. It’s a stable, human-level dexterous teleoperation system and a suite of autonomous policies trained with it. Pen spinning, complex in-hand reorientation, and long-horizon tool use—once seen as the holy grail of manipulation—are now unlocked. 🧵👇 The hardware is already here; we have some incredible high-DoF robotic hands. The bottleneck? The controller. Most current systems are stuck in "quasi-static" grasping mode. Meanwhile, dynamic in-hand dexterity has remained severely limited. 🧠 To unlock the massive capabilities of human-like hands, we need to build an excellent "cerebellum" for dexterous hands. TeleDexter solves this with a novel co-tracking approach: it simultaneously tracks both human hand kinematics and object states, beautifully bridging the gap between human intent and robotic control. In order to train a better co-tracking policy that works robustly in the real world, we designed : (1) a hybrid reward design that combines consecutive goal reaching and dense tracking, (2) an action masking strategy during training that enhances sim2real performance, (3) a dexterous curriculum for learning the long-horizon interactions. Each design is inspired by numerous trials and countless real-world experiments. We’ve synthesized all the system details, engineering challenges, and core insights into our latest post. If you're interested in the future of dexterous manipulation, grab a coffee and check it out (9-min read): If you have more time, check out the paper:

Siyuan Huang

12,584 views • 2 months ago