Загрузка видео...

Не удалось загрузить видео

На главную

Train in sim. Run it for real. Teach it new tricks. 🦆 This is the sim2real loop that powers Microduck. RL stack open source here : Buy it here : Git : Join our community :

pollenrobotics's profile picture

Pollen Robotics

19,712 subscribers

326,836 просмотров • 1 месяц назад •via X (Twitter)

Наука и технологии
LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 32

Фото профиля doogie
doogie1 месяц назад

play dead!

Фото профиля DR.David
DR.David1 месяц назад

Is the listed price €340 VAT-inclusive or exclusive? Your Shipping Policy states VAT is included in the shown price for EU orders, but the checkout is adding €74 VAT on top.

Фото профиля Lucas Cooper-Bey
Lucas Cooper-Bey1 месяц назад

@ClementDelangue Can’t believe gta6 extended look came out today and im way more excited for my Microduck

Фото профиля thawingfrog
thawingfrog1 месяц назад

Graphite Redesign 😬

Фото профиля Bart S.
Bart S.1 месяц назад

Hey @pollenrobotics, how long Microduck can live on a single battery charge?

Фото профиля Sebastian S. Cocioba🪄🌷
Sebastian S. Cocioba🪄🌷1 месяц назад

You guys are wonderful. Please stay true to this path so others can point when pitching and an evosystem of good follows.

Фото профиля Lben
Lben1 месяц назад

How do we train it? Any tutorials?

Фото профиля Haryn
Haryn1 месяц назад

I think I am gonna train bro to follow me and only me ! And we go travel ! Camping… scan for wild life? Holy sh$$ how far can we train this guy?

Фото профиля Thijs van Schadewijk
Thijs van Schadewijk1 месяц назад

I will need to train my duck to feed my dog. If she does that they will get along well.

Фото профиля Johannes Tscharn
Johannes Tscharn1 месяц назад

The biggest reason for me to buy this!

Фото профиля Naka
Naka1 месяц назад

hey what age group do you think this is sensible for? I showed one to my son, but I don't know if he is old enough

Фото профиля Neeraj Kumar
Neeraj Kumar1 месяц назад

Recovery after transfer is the useful benchmark. How wide is the sim to real gap across surfaces?

Фото профиля LightOrigins
LightOrigins1 месяц назад

Want Microduck to go somewhere? Just tell it. Our open-source navigation brain LightNav-0 now runs on it. Try it :)

Фото профиля vini
vini1 месяц назад

@clarinhatech

Фото профиля Kevin
Kevin1 месяц назад

looks so cool

Фото профиля Sagar
Sagar1 месяц назад

Sim2real is where robotics stopped being a demo. Open sourcing the RL stack is the real move.

Фото профиля Angel
Angel1 месяц назад

World changing moment.

Фото профиля Indivar Mishra
Indivar Mishra1 месяц назад

I want one so bad!

Фото профиля Devil Cloud
Devil Cloud1 месяц назад

Awesome work to all of you. Should definitely get some glow in the dark panel options.

Фото профиля Sam Snelling
Sam Snelling1 месяц назад

Assembly? I see nothing on the site about assembly and reachy was all diy

Фото профиля Easton3D
Easton3D1 месяц назад

Insane launch! I’m building a whole lineup of robotic characters for my book series B.U.R.T.Y. - all 3D‑designed, no robotics (yet!). Would love to see my characters come to life like this one day.

Фото профиля FieryStardust
FieryStardust1 месяц назад

I teach typing, coding, 3d printing to K-6th grade students. They love all robots and do different projects where they have to design a robot that helps a specific person, group, or the planet and I can’t wait to show them your robots! Our kids need more classroom robots!

Фото профиля Sully
Sully1 месяц назад

is data all stored locally?

Фото профиля Y is in the house
Y is in the house1 месяц назад

can it be trained to respond to voice commands?

Фото профиля मो क्ष
मो क्ष1 месяц назад

tysmmmm

Фото профиля Chirag
Chirag1 месяц назад

India distributor?

Фото профиля Samantha
Samantha1 месяц назад

Hey Pollen Robotics! Your open-source humanoid robotics work caught our eye. We’re exploring how accessible platforms can accelerate real-world robotics and would value your perspective. DM us if you’re open to a short interview chat.

Фото профиля Jeff Morris
Jeff Morris1 месяц назад

Why no Australia. Existing reachy mini customer.....

Фото профиля wek up neo
wek up neo1 месяц назад

wow simply genius and love the design

Фото профиля HAB1B0 German Raffleking 🇩🇪🇹🇷👑{🦅}
HAB1B0 German Raffleking 🇩🇪🇹🇷👑{🦅}1 месяц назад

Hey @pollenrobotics do you also ship to Germany?

Фото профиля jackyyzm
jackyyzm1 месяц назад

Hello bro ，Are you the inventor of Microduck? Nice to meet you. I want to know how to sell this product in China.

Фото профиля piggie
piggie1 месяц назад

can we add a microphone/speaker to it?

Похожие видео

We built a small biped robot you can teach new tricks to. Train it in simulation, run it on the real thing. Meet Microduck 🦆 $399, shipping before Christmas.
0:51

We built a small biped robot you can teach new tricks to. Train it in simulation, run it on the real thing. Meet Microduck 🦆 $399, shipping before Christmas.

Pollen Robotics

1,640,849 просмотров • 1 месяц назад

waiting for the microduck robot to ship? 🦆 play with it on the microduck simulator space, running the same RL policies as the real robot ▶️
0:21

waiting for the microduck robot to ship? 🦆 play with it on the microduck simulator space, running the same RL policies as the real robot ▶️

Hugging Apps

28,486 просмотров • 1 месяц назад

🤖WILD: This $399 robot duck is selling one unit every four seconds. Hugging Face just launched the "Microduck," an open-source robot anyone can teach new tricks. It can waddle, roller skate, pick things up with its beak, and get back up when it falls.
0:51

🤖WILD: This $399 robot duck is selling one unit every four seconds. Hugging Face just launched the "Microduck," an open-source robot anyone can teach new tricks. It can waddle, roller skate, pick things up with its beak, and get back up when it falls.

Coin Bureau

130,745 просмотров • 1 месяц назад

Prime Intellect engineers explained how they train reasoning models over the open internet in 30 minutes - better than $3000 distributed training courses. split policy and rollouts across nodes -> run agents in parallel envs -> verify with LLM judges -> gradient-sync over the internet -> train Llama, Qwen, Gemma at cluster scale on rented GPUs. That loop is why open reasoning models are catching closed labs without owning a data center. Prime-RL + verifiers + distributed rollouts + LLM judges + multi-cloud GPUs - that's the stack. Watch and save it, then launch your first distributed RL run this week.
30:09

Prime Intellect engineers explained how they train reasoning models over the open internet in 30 minutes - better than $3000 distributed training courses. split policy and rollouts across nodes -> run agents in parallel envs -> verify with LLM judges -> gradient-sync over the internet -> train Llama, Qwen, Gemma at cluster scale on rented GPUs. That loop is why open reasoning models are catching closed labs without owning a data center. Prime-RL + verifiers + distributed rollouts + LLM judges + multi-cloud GPUs - that's the stack. Watch and save it, then launch your first distributed RL run this week.

h100envy

40,022 просмотров • 3 месяцев назад

Giving Microduck object-tracking vision capabilities! 🦆👀🤖 It can now play fetch and doesn't get confused by other objects. Using Roboflow Trackers library I made Microduck keep track of the different objects in its sight and has a control policy for reaching the ball it tracks. Open-Source code below 👇
0:28

Giving Microduck object-tracking vision capabilities! 🦆👀🤖 It can now play fetch and doesn't get confused by other objects. Using Roboflow Trackers library I made Microduck keep track of the different objects in its sight and has a control policy for reaching the ball it tracks. Open-Source code below 👇

Alex Bodner

31,551 просмотров • 1 месяц назад

Here is the video for our new single please presave it here
4:03

Here is the video for our new single please presave it here

Calling All Astronauts

33,613 просмотров • 1 год назад

Our new sponsor SoundMoney is bringing gold to the people. Buy real gold and silver from $1. Store it in America. Sell it with one click. Support our work and get your FREE account here:
0:24

Our new sponsor SoundMoney is bringing gold to the people. Buy real gold and silver from $1. Store it in America. Sell it with one click. Support our work and get your FREE account here:

The Post Millennial

69,735 просмотров • 2 месяцев назад

Launching SYNTHETIC-2: our next-gen open reasoning dataset and planetary-scale synthetic data generation run. Powered by our P2P inference stack and DeepSeek-R1-0528, it verifies traces for the hardest RL tasks. Contribute towards AGI via open, permissionless compute.
1:27

Launching SYNTHETIC-2: our next-gen open reasoning dataset and planetary-scale synthetic data generation run. Powered by our P2P inference stack and DeepSeek-R1-0528, it verifies traces for the hardest RL tasks. Contribute towards AGI via open, permissionless compute.

Prime Intellect

360,859 просмотров • 1 год назад

opus 5.5 is something else.. so many people asked me to open source the 3d water from my last post. a few even wanted to buy it so here it is. my first open source drop code » demo » show me what you make with it
0:26

opus 5.5 is something else.. so many people asked me to open source the 3d water from my last post. a few even wanted to buy it so here it is. my first open source drop code » demo » show me what you make with it

Aurelien

256,111 просмотров • 22 дней назад

This guy explained agent harness, memory, and loop engineering in one video - clearer than most paid courses a raw LLM with no harness is a wild horse. all power, no direction > trace the run -> score it with an LLM -> diagnose the bottleneck -> patch it -> ship it that loop is what lets an agent get sharper on its own, run after run memory + harness + loop engineering + evals - that's the whole stack in one picture watch it, then grab the diagram below
17:43

This guy explained agent harness, memory, and loop engineering in one video - clearer than most paid courses a raw LLM with no harness is a wild horse. all power, no direction > trace the run -> score it with an LLM -> diagnose the bottleneck -> patch it -> ship it that loop is what lets an agent get sharper on its own, run after run memory + harness + loop engineering + evals - that's the whole stack in one picture watch it, then grab the diagram below

Mr. Buzzoni

45,204 просмотров • 2 месяцев назад

AI is concentrating, in a few countries, a few companies, a few chips. Open-source AI is the only real check on that. And it only works if the infrastructure to run it actually exists. Today, we're announcing our Series A, $20 M, to make this a reality.
1:16

AI is concentrating, in a few countries, a few companies, a few chips. Open-source AI is the only real check on that. And it only works if the infrastructure to run it actually exists. Today, we're announcing our Series A, $20 M, to make this a reality.

Eugene Cheah - AI builder @ 🇸🇬|🇺🇸

31,277 просмотров • 5 месяцев назад

The Pact is here, and with it comes our new open-world activity Retaliation. Here's a peek at how it works!
1:52

The Pact is here, and with it comes our new open-world activity Retaliation. Here's a peek at how it works!

Tom Clancy's The Division

49,243 просмотров • 1 год назад

Historically, RL policies for robots have been trained in synthetic, untextured environments, limiting perceptive policies to depth images where the sim-to-real gap is manageable. RGB has always had more potential, but leveraging it to train policies in simulation remained an open problem. Partnering with Niantic Spatial 🌎 and NVIDIA Robotics, we built a pipeline that addresses exactly that. We can now scan a real deployment site with off-the-shelf hardware, reconstruct it into a photorealistic Gaussian splat, and run massively parallel RL training. The policies trained in our Gym environment then transfer zero-shot to the real robot and environments they were trained for. This enables faster deployment of more capable and robust policies for the end user. The new resulting capabilities are a big step towards solving sim-to-real and also apply well beyond navigation. Read the full technical breakdown on our blog; link in the comments. #HumanoidRobots #Flexion #NianticSpatial #NVIDIA
1:21

Historically, RL policies for robots have been trained in synthetic, untextured environments, limiting perceptive policies to depth images where the sim-to-real gap is manageable. RGB has always had more potential, but leveraging it to train policies in simulation remained an open problem. Partnering with Niantic Spatial 🌎 and NVIDIA Robotics, we built a pipeline that addresses exactly that. We can now scan a real deployment site with off-the-shelf hardware, reconstruct it into a photorealistic Gaussian splat, and run massively parallel RL training. The policies trained in our Gym environment then transfer zero-shot to the real robot and environments they were trained for. This enables faster deployment of more capable and robust policies for the end user. The new resulting capabilities are a big step towards solving sim-to-real and also apply well beyond navigation. Read the full technical breakdown on our blog; link in the comments. #HumanoidRobots #Flexion #NianticSpatial #NVIDIA

Flexion

73,099 просмотров • 2 месяцев назад

I rebuilt most of OpenClaw's core in a single workflow: - 25 blocks - 29 connections - Short + long-term memory - Multi-channel (Telegram + Slack) Didn't build it manually. Stack is fully open-source. Self-host, run local models, own it end-to-end. Full walkthrough: Chapters: 00:00 - Intro 01:00 - SimClaw in action: planning my day, finding meetings, sending email 04:05 - Long-term memory capability 05:52 - Inside the workflow: how it's wired 12:09 - The plot twist 12:50 - Building an entire workflow using a single prompt 15:42 - Why this is an OS for your AI workforce 17:00 - Try it yourself If you want to see the open-source stack that powers all of this, check out Sim on GitHub and drop a star if you find it useful:
17:21

I rebuilt most of OpenClaw's core in a single workflow: - 25 blocks - 29 connections - Short + long-term memory - Multi-channel (Telegram + Slack) Didn't build it manually. Stack is fully open-source. Self-host, run local models, own it end-to-end. Full walkthrough: Chapters: 00:00 - Intro 01:00 - SimClaw in action: planning my day, finding meetings, sending email 04:05 - Long-term memory capability 05:52 - Inside the workflow: how it's wired 12:09 - The plot twist 12:50 - Building an entire workflow using a single prompt 15:42 - Why this is an OS for your AI workforce 17:00 - Try it yourself If you want to see the open-source stack that powers all of this, check out Sim on GitHub and drop a star if you find it useful:

Akshay 🚀

65,756 просмотров • 5 месяцев назад

The coolest part of Microduck might not be the robot. It's how it learns to move. This repo shows the full reinforcement learning pipeline behind Microduck. The robot learns in simulation using MuJoCo, PPO and GPU-based training, then the trained policy is exported to ONNX and deployed on the real robot. But the real challenge is sim-to-real. The simulation models things like actuator behavior, battery voltage, command delays, friction and even gear backlash. That helps make the learned policy work when it moves from simulation to the physical robot. The repo includes policies for walking, fall recovery, sitting, standing, kicking, ground picking and even roller skating. Train in simulation. Test the behavior. Export the policy. Run it on a real robot.
1:01

The coolest part of Microduck might not be the robot. It's how it learns to move. This repo shows the full reinforcement learning pipeline behind Microduck. The robot learns in simulation using MuJoCo, PPO and GPU-based training, then the trained policy is exported to ONNX and deployed on the real robot. But the real challenge is sim-to-real. The simulation models things like actuator behavior, battery voltage, command delays, friction and even gear backlash. That helps make the learned policy work when it moves from simulation to the physical robot. The repo includes policies for walking, fall recovery, sitting, standing, kicking, ground picking and even roller skating. Train in simulation. Test the behavior. Export the policy. Run it on a real robot.

Praveen Kumar Verma

138,017 просмотров • 1 месяц назад

also check out our code here we haven't released everything yet but there's enough there that you can train a policy, eval it in sim, and view the rollouts live in viser!
0:50

also check out our code here we haven't released everything yet but there's enough there that you can train a policy, eval it in sim, and view the rollouts live in viser!

Arthur Allshire

15,716 просмотров • 4 месяцев назад

The discussion of the rapture of the church is a source of joy for us, as it reaffirms that our place isn't here on Earth. This brings comfort to our hearts.
1:11

The discussion of the rapture of the church is a source of joy for us, as it reaffirms that our place isn't here on Earth. This brings comfort to our hearts.

Huiothesia

10,339 просмотров • 3 лет назад