Xiaomi drops XR-1, a robot foundation model pre-trained on 100K hours of embodiment-free data

Xiaomi-Robotics-1

Xiaomi Robotics open-sourced XR-1, a ready-to-use robot foundation model. It pre-trains on 100K hours of embodiment-free manipulation videos across 1,700+ scenarios, then post-trains on 7,200 hours of real-robot data for embodiment and instruction alignment. Pre-training shows clean scaling laws—lower action error with more data and larger models—and those gains transfer directly to real-robot success rates with no sign of saturation yet. After post-training, XR-1 picks up new tasks like phone packing and printer refilling from under 10 hours of demos on average, hitting 75% overall success (nearly 2× π 0.5); with under 40 hours it reaches 85%. It also achieves SOTA on four sim benchmarks. Code, weights, and paper are public.

Why it matters: Xiaomi Robotics open-sourced XR-1, a robot foundation model with code, weights, and paper. The two-stage recipe (100K hrs embodiment-free pretraining + 7,200 hrs real-robot post-training) and the pretraining scaling law are hard signals, directly comparable to π 0.5. Scored as...