Robot AI model

UnifoLM-VLA-0

UnifoLM-VLA-0 is Unitree's language-conditioned model for humanoid manipulation. Its release includes continued visual-language pretraining, robot-action training, and a checkpoint fine-tuned on Unitree's open manipulation datasets.

Records last verified Oct 13, 2026 · Send a correction

Organization
Unitree Robotics
Type
VLA
Released / introduced
Jan 29, 2026
Public weights
Public weights
Licence / scope
CC BY-NC-SA 4.0
Parameters
Not disclosed
Base model
UnifoLM-VLM-Base
Humanoid association
Humanoid-native
Inputs
vision, language, proprioception
Outputs
robot actions

Release dates can mark an announcement or paper introduction. Public checkpoints, code licences and weight terms are distinct; follow the publisher’s links for the stated scope.

Documented embodiments

Source-associated hardware; this list does not establish plug-and-play compatibility or commercial robot capability.

  • Unitree G1 →

    The official README identifies G1 block-stacking and bag-insertion examples.

Disclosed training data

  • Unitree open G1 manipulation datasets
  • General image-text VQA and open robot datasets

More robot AI models

Evidence

Sources

  1. [s1] UnifoLM-VLA-0 official README and release date ↗

    Unitree Robotics · official · accessed Oct 13, 2026

  2. [s2] UnifoLM-VLA-Base model card ↗

    Unitree Robotics · official · accessed Oct 13, 2026