Robot AI model
GO-1
GO-1 is the AgiBot World team's VLA pretrained on its robot manipulation dataset. The public model uses an InternVL 2.5 backbone; the open-release study examines training and evaluation practices using AgiBot G1.
Records last verified Oct 13, 2026 · Send a correction
- Organization
- AgiBot World team / OpenDriveLab
- Type
- VLA
- Released / introduced
- Sep 19, 2025
- Public weights
- Public weights
- Licence / scope
- CC BY-NC-SA 4.0
- Parameters
- Not disclosed
- Base model
- InternVL 2.5-2B
- Humanoid association
- Humanoid demonstrated
- Inputs
- vision, language
- 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.
- AgiBot G1 →
The repository identifies the G1 as the Genie Studio platform for GO-1.
Disclosed training data
More robot AI models
Evidence
Sources
- [s1] GO-1 official model card ↗
AgiBot World team · official · accessed Oct 13, 2026
- [s2] AgiBot World repository: GO-1 release and Genie Studio ↗
AgiBot World team / OpenDriveLab · official · accessed Oct 13, 2026
- [s3] Open-sourcing GO-1: building VLA systems at scale ↗
OpenDriveLab · official · published Sep 19, 2025 · accessed Oct 13, 2026