Many people dont know but Mujoco cpu and MJWarp do have different implementations. The main difference we identified is how the engines represent contact between the cylindrical can and the fingers. At the exact same hand and can pose, CPU Mujoco generated 15 hand can contact points, versus 4 in the original MJWarp implementation. Those points determine the forces and friction supporting the grip. So TLDR, we asked astra to make an expert video which looked successful but we didnt know that it used normal Mujoco so our MJWarp in sim could never perform the action correctly.

Join us for an in-person Q&A with the authors of MuJoCo, Tom Erez & Yuval Tassa from @GoogleDeepMind at the Google booth (#411#) today at 3PM! Bring your questions. #ICLR2026# #MuJoCo#

the duck learnt to sit and stand on command. 15,000 iterations of PPO in 4h32min on my RTX 5090, 4096 MuJoCo environments in parallel. mean reward 1.7 at iteration 0, 133 at the end. in the clip: stand at 116 mm, sit at 59 mm, each switch under a second, and it holds the pose until the next command. third skill in the tree after walking and getting up!

$28 3D printed teleoperator device. And of course it's open source! The design features clean cable routing, same resting position as the SO-ARM101 and optional rubber bands that improve teleoperation experience (great idea). Used with MuJoCo simulator in the video below but can also be used with the follower arm. Awesome contribution from our community member! Check out the project here:

JUST FOUND OUT FABLE IS CAPABLE OF DESIGNING + 3D PRINTING WORKING FIDGET TOYS LOL In only a couple prompts I had Fable - Design fidget toy variations - Test the physics in MuJoCo & update based on feedback - Make it REAL with a 3D Printer (Bambu X1C) FULL BREAKDOWN HERE:

Hours of manual asset authoring => Down to 1–2 minutes with 1 prompt. SR Platform generates interactive, simulation-ready 3D assets for training robotics, complete with articulated objects and physics. Native support for MuJoCo and NVIDIA Isaac Sim. Coming in July!

My essay at TheSequence today covers the topic of environments in AI post-training AI environments are becoming a core ingredient of model capability: executable worlds that generate trajectories and feedback, not just exams. This essay traces their path from Atari, OpenAI Gym, and AlphaZero through MuJoCo to the RLVR-driven post-training behind Tulu 3 and DeepSeek-R1, then surveys the new generation of agent environments, including SWE-Gym, WebArena, OSWorld, TheAgentCompany, Tau2-bench, and AgentMercury. Check it out:

Can humanoids recover in interaction tasks? 📦🤖 Excited to share I-BFM! Reward-conditioned policy: carry, push, kick. FB: robot–object–contact dynamics; LOGO: contact now → task goal later. G1 MuJoCo Carry: 94.3% → 89.3% after robot disturbance. 🌐

Build a humanoid robot yourself! 🪚 OpenArm is an open-source humanoid robot.  It comes with full CAD, control code, firmware, and simulation tools, everything needed to build, modify, and operate it. The arms are designed to be compliant and backdrivable. Teleoperation is supported, with force feedback and real-time gravity compensation so operators can guide the arm naturally. Super important, on the simulation side, OpenArm works with platforms like MuJoCo and Isaac Sim, letting developers test policies in virtual environments before running on hardware.  Assemble it yourself from a kit or get it prebuilt, the goal is accessibility for research labs, small teams, and enthusiasts. The project is run by @enactic_ai in Tokyo, Japan, and aims to lower the barrier for experimenting with dexterous manipulation. Let's put robotics in mainstream! 🔥🔥 🔗 Find the link here: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →