Frame Robotics

No install. No GPU. Frame Sim runs eleven robots on MuJoCo physics in your browser, with real policies pulled straight from Hugging Face: LeRobot ACT handing a cube across two ALOHA arms, Unitree Go2 and G1 walking on their published controllers. Pick a table, a factory or a farm, tell the robot what to do, then record the run, replay it and compare.

We took robots into an orchard. How do you think it went? 🍏

We pay people to film the work they already do. @KristofSzabo_ walks through the whole FRAME Capture flow: how to pick a task, film it, and submit. Ready to earn? Learn more: framerobotics.ai/capture

Meet Lynx M20: a new set of eyes for agriculture. With our client, we’re exploring how computer vision can help robots count cattle. Farmers can fully delegate herd monitoring and fence inspection, helping them see more and spot problems sooner. Is Black Mirror here already?

We spend a lot of the robot-teaching time in two places: real video, and simulation. What you are seeing is the real action, then the same task in simulation on our robot. A person gets through the motion much faster than we can teleoperate it, which is why we keep filming real work instead of living on robot hours. The long part is everything between the two shots. We take the human motion from the video, put it on our joints, and run it in sim until the robot is actually doing it. A motion that looks right on the person does not land on our arm until we have worked it over. Sim is where that happens, because we can repeat it all day before we put it on hardware. We are done with a clip when the arm in sim can do what the person did. Filming it is just the start.

Can a robot become an animal supervisor? We’re about to find out.

A factory robot works because the product never changes. An orchard breaks that assumption completely. Different size, different angle, different height, half hidden behind leaves. That's why real-world footage is the bottleneck.