Skip to content
embodied.guide

Back to the AI models

AgibotChina2026

GO-2

An AI model: a vision-language-action model (VLA) takes camera images and an instruction and turns them into movements for a robot. This page describes the software, not a robot.

Robotsnot stated

Key figures as stated by the developer

0 of 6 stated · 6 gaps

The manufacturer states none of the 6 key figures.

How to read the figures
14kgA30.9lbA
The letter in the grey square after a figure points to the source the figure came from. Robots with several sources carry several letters.
74.5lb
Imperial figures are our own conversion unless the manufacturer states an imperial figure itself. The source letter always belongs to the metric figure — a conversion has no source of its own.

The four states

Versions and notes in the key figures (3)

Model type

Note The developer calls it something else and does not label it vision-language-action or world action model: "its next-generation foundation model for embodied AI".

Controls

Note AGIBOT on System 1: "it continuously receives high-level intents and combines them with real-time observations to generate specific control signals".

Inputs

Note AGIBOT on System 1: "it continuously receives high-level intents and combines them with real-time observations to generate specific control signals".

Full specifications

1 of 12 fields stated
The letter points to the source at the foot of the page. Fields without a figure are dashed: the developer does not state them — that is not the same as zero, and not the same as no.
Field Value Source and caveats
Input and output
Model typenot stated

Note The developer calls it something else and does not label it vision-language-action or world action model: "its next-generation foundation model for embodied AI". ⁠A

Inputsnot stated

Note AGIBOT on System 1: "it continuously receives high-level intents and combines them with real-time observations to generate specific control signals". ⁠A

Controlsnot stated

Note AGIBOT on System 1: "it continuously receives high-level intents and combines them with real-time observations to generate specific control signals". ⁠A

Degrees of freedom controllednot stated
Size and speed
Parametersnot stated
Control ratenot stated

Note Two layers without numbers: "Semantic Planning Module (System 2): Operates at a lower frequency." "Action Following Module (System 1): Operates at a higher frequency." ⁠A
Where it runs and what it is built on
Runs wherenot stated

Note AGIBOT names "Cloud-based collaborative training" for its deployment loop, not where GO-2 runs. ⁠A

Base modelnot stated
Openness
Accessnot stated
Licensenot stated
The developer’s own claims
Learns new tasksneeds retrainingA

Developer claim: GO-2 uses a "pre-training + post-training + data feedback loop paradigm", with "Online post-training in real-world environments".

Works across robotsnot stated

Note AGIBOT's claim concerns many units: "Continuous data collection across fleets of robots". ⁠A

Yes only if the developer itself says the model runs on different robot models, for example robots from two manufacturers or two of one manufacturer’s models. Several units of the same robot do not count.

Notes

  • AGIBOT describes GO-2 as a unified architecture that integrates logical reasoning and action execution within a single system.
  • GO-2 has two layers: a lower-frequency planning module (System 2) and a higher-frequency action-following module (System 1).

Sources

One line per source, not per figure: a source has one retrieval date. The letter beside a figure points to the line it came from.