Skip to content
RobothecaBeta
Log in

MiniCPM-Robot (RobotManip, RobotTrack)

  • AI model
  • Humanoids and robot learning

MiniCPM-Robot is a family of on-device embodied AI models for robots, with a manipulation model and a target-tracking model that run locally on robot hardware.

Made by OpenBMB.

Source on GitHub: github.com/OpenBMB/MiniCPM-Robot (opens in a new tab). Facts read at commit f666593 (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 f666593 (opens in a new tab), on 16 Oct 2026.

Commercial use
Yes

What it does

  • MiniCPM-RobotManip is a 1.5B vision-language-action model for robot manipulation, using one set of weights across tasks.
  • It supports streaming inference that adds historical observations to the model context, up to 1 minute of visual memory.
  • MiniCPM-RobotTrack is a 0.9B model for embodied target tracking that works from vision and natural-language commands alone.
  • The tracking model is deployed on a Unitree Go2 EDU with a Jetson Orin NX, reaching 5+ FPS and about 180 ms end-to-end latency.
  • Evaluation integrations exist for LIBERO, CALVIN, RoboTwin2 and RMBench, and a Go2 deployment guide covers the real-robot workflow with a dry-run default.

Written by Robotheca from the project's README (opens in a new tab) at commit f666593.

Getting started

Difficulty
Advanced

Maintenance

Status
Active: a commit in the last 90 days
Newest commit
25 Aug 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.