co-founder and robotics guy @ mimic | ETH Zürich
Zurich, Switzerland
Joined July 2024
- Tweets52
- Following438
- Followers119
- Likes350
two days of continuous dexterous manipulation at @NVIDIAGTC . zero downtime, zero vaporware, zero bs.
git gud
Visit the @mimicrobotics x @bfl_ai booth at GTC Berlin - but only if you’ve filled your vaporware quota for the day. A shop floor ready stack with end to end autonomous dexterity. lfg team!
The team Just got set up at GTC Berlin. You should come over to the @mimicrobotics x @bfl_ai booth, it's really good!
Benedek Forrai retweeted
(part 2) I asked people: "which robotics labs and startups have the highest talent density?"
Full list: frontierlist.com/robotics-ta…
1 vote each:
- Amazon FAR (@pabbeel, @peterxichen, @rocky_duan)
- Antim Labs (@Shreyko, Viswajit Vinod Nair)
- Bedrock Robotics (@bsofman, @kevinmpeterson1, Ajay Gummalla, Tom Eliaz)
- Build AI (@patrickrim, @eddybuild, @jonathanjia05, @zikangjiang)
- Dimensional (@stash_pomichter)
- Dream Labs (@jang_yoel + 3 in stealth)
- Enact (@jamesw_stevens, @govindchada)
- Eye Candy Robotics (@Mankaran32, @_raghuvamsi, Alqama Shaikh)
- Foundry Robotics (@adarshkulkarni)
- GDM Robotics (@parada_car88104)
- Hark (@adcock_brett)
- Medra (@michellearning)
- microagi (@bercankilic, @YoanIlievX, @antonwontonnn, Nico Nussbaum, Artjem Weissbeck)
- Mimic (@stefan_weirich, @elvisnavah, @GravertStephan, @forraibotics)
- NVIDIA GEAR Lab (@DrJimFan, @yukez)
- Pantograph (@apagajewski, @_kelsey_pool)
- Reward AI (@zipengfu, @chenwang_j)
- RobbyAnt (Xing Zhu, Yujun Shen, Yongtao Huang)
- Runway (GWM) (@c_valenzuelab, @agermanidis, @matamalaortiz)
- Somana
- Trener Robotics (Asad Tirmizi, Lars Tingelstad)
- Ultra (@JonMSchwartz, Max Friefeld, @chetan_, Oliver Ortlieb)
- Unitree (Wang Xingxing)
- Vision Lab (@JamesKuj_OG, Zhichu Ren, @zwbgood6)
- Walden Robotics (@RussTedrake, @Ben_Burchfiel, Siyuan Feng, @RaresAmbrus, @adnothing, Kerri Fetzer-Borelli, Dave Johnson)
- Watney Robotics (@rgannon_, Sean Cheong)
- Xaba (Massimiliano "Max" Moruzzi)
empirical take: the xp juniors bring in robotics has become very bimodal: you’re either goated or cooked. even in good unis like eth.
together with @ethroboticsclub we’re running a lecture series to even the playing field. sign up here luma.com/23dsvh0f
i would be surprised if anyone in multimodal/robot learning would be impressed by these nothingburger demos
this is a cute experiment to run but thats about it
I am seeing claims around LLMs, specifically Astra having made serious progress on robotics. The tasks that are demonstrated are simple pick and place tasks with parallel jaw grippers. LLMs can do planning, and the impressive demos in these tasks primarily show that. But robotics is also about dexterity and dynamics - which is why we need high frequency controllers/policies that can deal with torques and forces. So here is a simple challenge. Can you prompt an LLM to output the high frequency control commands for a legged robot in varying terrain e.g. ashish-kmr.github.io/ RSS 2021, CoRL 2022 (this is by now 5 year old technology, so I am not picking a particularly hard task). I am not questioning the usefulness of LLMs for high level planning or in agentically assisting a robotics researcher (we use them all the time!).
>most on the frontier figured out icl already
my sides: in orbit
the special type of models worth to fight the borrow-checker for: today, @mimicrobotics is launching FLUX-mimic, co-built with @bfl_ai ‘s FLUX 3. check out our blog: mimicrobotics.com/blog/intro…
Benedek Forrai retweeted
I remember when we started working on this, it seemed so far-fetched. Now FLUX-mimic, built together with @bfl_ml, is being tested and deployed with @AudiOfficial
Benedek Forrai retweeted
Introducing FLUX-mimic, a next-generation Video-Action Model for general purpose dexterity, developed in partnership with @bfl_ai.
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, on complex, multi-step manipulation long considered impossible for conventional automation.
Benedek Forrai retweeted
Stat monster
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, building every layer ourselves around one fixed point, the human hand.
The M1 is a highly backdrivable, tendon-driven hand that covers the full range of human capability, from heavy payloads to fine manipulation.
(maybe not so) hot take: robotics is also a sw infra problem, not “just” a learning/data/hardware problem. Today, we’ve shared a glimpse into our take: truly zero copy, no serialization/deserialization. This is what it unlocks: mimicrobotics.com/blog/solvi…
All rust btw
Benedek Forrai retweeted
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, building every layer ourselves around one fixed point, the human hand.
The M1 is a highly backdrivable, tendon-driven hand that covers the full range of human capability, from heavy payloads to fine manipulation.
Benedek Forrai retweeted
A few months ago, I travelled to Zurich together with @NVIDIArobotics to spend time inside one of Europe's most exciting robotics startups...
@mimicrobotics!
Their mission? To solve one of the hardest problems in robotics: human-level dexterity.
Because before robots can replace physical work, they first need to master the human hand.
In this episode, we go behind the scenes with the team building AI-powered robotic manipulation, discuss why dexterity is still one of the biggest bottlenecks in automation, and explore what it will take to bring truly capable robots into factories.
Mimic Robotics is part of the NVIDIA Inception, a program for startups and VCs, and is leveraging Cosmos to advance AI-powered robot learning.
If you're interested in Physical AI, robotic manipulation, and where this industry is actually heading, I think you'll enjoy this one.
Timestamps:
0:00 Inside Mimic Robotics
0:24 Building the future of robot manipulation
1:19 The mission behind Mimic Robotics
1:35 The AI powering autonomous robots
2:14 Why Mimic chose NVIDIA Cosmos
2:34 Video foundation models vs. traditional robotics
2:46 Building a multidisciplinary robotics team
3:24 Bridging customers and cutting-edge AI
4:40 Why Zurich is the perfect robotics hub
5:20 Europe vs. America: Scaling a robotics startup
6:30 Bringing generative AI to factory robots
6:49 The race to build Europe's robotics hyperscaler
♻️ Know someone working in robotics? They'll probably want to see this.