Skip to content
RobothecaBeta
Log in

Xiaomi-Robotics-1 and Xiaomi-Robotics-0

  • AI model
  • Humanoids and robot learning

A vision-language-action robot foundation model pre-trained on over 100K hours of embodiment-free UMI trajectories, then post-trained for mobile manipulation in unseen environments.

Made by Xiaomi Robotics, China (checked on github.com/XiaomiRobotics (opens in a new tab)).

Source on GitHub: github.com/XiaomiRobotics/Xiaomi-Robotics-1 (opens in a new tab). Facts read at commit 0dd7aef (opens in a new tab).

Licenses

Code
Apache-2.0 (opens in a new tab)

Read by Robotheca from LICENSE (opens in a new tab) (Apache-2.0) at commit 0dd7aef (opens in a new tab), on 16 Oct 2026.

Commercial use
Yes

What it does

  • Couples a pre-trained Qwen3-VL vision-language model with a Diffusion-Transformer action head via a Mixture-of-Transformers design.
  • Pre-trained on embodiment-free UMI data, with a VLM-based auto-labeling pipeline that annotates trajectory clips.
  • Post-trained on over 10K hours of cross-embodiment data to align actions to real robots and natural-language instructions.
  • Aims for out-of-the-box mobile manipulation in unseen environments and adaptation to new tasks from a few hours of demonstrations.
  • Reports state-of-the-art results on RoboCasa, RoboCasa365, VLABench and RoboDojo, with fine-tuned checkpoints for each.
  • Code and checkpoints for post-training, inference and benchmark evaluation were released, and the model is available on Hugging Face.

Written by Robotheca from the project's README (opens in a new tab) at commit 0dd7aef.

Getting started

Difficulty
Advanced

Maintenance

Status
Active: a commit in the last 90 days
Newest commit
9 Sept 2026 on main
Latest release
None published on GitHub

Checked on GitHub on 16 Oct 2026.

Links

Robots on Robotheca that use it

No robot on Robotheca is linked to it yet. Links come from a robot's robot.yaml, from its dependency files read as text, or from Robotheca's editors.

Robotheca lists AI models as a reference: nothing is hosted or run here, and every fact links to where it was read.