Skip to content
RobothecaBeta
Log in

GigaBrain-0

  • AI model
  • Humanoids and robot learning

A vision-language-action foundation model for generalist robot manipulation, trained on about 37,300 hours of heterogeneous embodied data and aimed at home and industrial tasks.

Made by GigaAI, China (checked on github.com/open-gigaai (opens in a new tab)).

Source on GitHub: github.com/open-gigaai/giga-brain-0 (opens in a new tab). Facts read at commit 5f450a9 (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 5f450a9 (opens in a new tab), on 16 Oct 2026.

Commercial use
Yes

What it does

  • Unifies understanding, prediction, and action through a three-system architecture.
  • Pretrained on 37.3k hours of real-robot, UMI, egocentric, simulation, and world-model-generated data.
  • Uses one-stage alignment training that jointly optimizes vision-language understanding and multi-embodiment action generation.
  • Provides training configs for AgileX PiPER and Maker H01 robots, with normalization-statistics and embodiment-ID setup.
  • Includes offline LeRobot rollout evaluation that reports per-episode MSE/MAE metrics.
  • Ships deployment servers and ROS clients for AgileX Cobot Magic and Maker H01, plus guides for RoboColiseum, RoboTwin 2.0, EBench, and VLM benchmarks.

Written by Robotheca from the project's README (opens in a new tab) at commit 5f450a9.

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.