yet another open-source eval rig, but it's my take

if you have followed me for the past few weeks, you’d see that I’ve been playing with a rather peculiar set up with the SO-101 and LeSlider:

this was nor just a random setup, this is what we at LiveKit have been designing, not for the purpose of my experiments, but to benchmark and debug LiveKit Portal.

today, we are open-sourcing it for the community, so everyone can have an easier time reproducing and share results at home.

here is the github.

i know this isn’t the first frame designed for stabling evals, but it’s ours, and it’s open, and it’s with a LeSlider!

along with this, we’ve released simulations files (URDF, MJCF, and USD), as well as a sample RL env in MJLab for pick-and-place tasks.

we want people to have fun, being close to the frontier as much as possible, and we want people to know it shouldn’t take thousands of dollars in equipment to reproduce some of the most amazing results the academic world has to offer.

personally, I’m going to deploy sim2real and real2sim2real pipelines on this soon, for fun, so look out for that.

if you have followed me for the past few weeks, you’d see that I’ve been playing with a rather peculiar set up with the SO-101 and LeSlider:

this was nor just a random setup, this is what we at LiveKit have been designing, not for the purpose of my experiments, but to benchmark and debug LiveKit Portal.

today, we are open-sourcing it for the community, so everyone can have an easier time reproducing and share results at home.

here is the github.

i know this isn’t the first frame designed for stabling evals, but it’s ours, and it’s open, and it’s with a LeSlider!

along with this, we’ve released simulations files (URDF, MJCF, and USD), as well as a sample RL env in MJLab for pick-and-place tasks.

personally, I’m going to deploy sim2real and real2sim2real pipelines on this soon, for fun, so look out for that.