Gemini Robotics 2 Introduction
Google DeepMind introduced Gemini Robotics 2 on July 30, 2026 as a vision-language-action model for intelligent whole-body robot control, alongside two additional models in the Gemini Robotics 2 family.
Historical classification
This verified record belongs to LINEAiGE's artificial-intelligence history layer beginning in 1956.
Why it matters
A dated first-party AI-system release milestone in embodied AI and robotics, documenting a model family designed to connect multimodal reasoning with robot motor control.
People, organizations & technologies
PEOPLE
Carolina Parada
ORGANIZATIONS
Google DeepMind
TECHNOLOGIES
Gemini Robotics 2
Evidence & verification
STATE
verified
CONFIDENCE
high
LAST VERIFIED
2026-09-21T04:07:00+08:00
Google DeepMind's first-party announcement is dated July 30, 2026 and explicitly says it is introducing Gemini Robotics 2. The announcement describes Gemini Robotics 2 as its most advanced vision-language-action model and says it converts vision and language input into motor control. This candidate records the dated introduction only and does not infer ancestry, influence, causality, development geography, universal priority, or relationship edges.
Sources
Gemini Robotics 2 brings whole body intelligence to robots
Google DeepMind
PUBLISHED
2026-07-30
Trace the chronology
Record identity
LINEAiGE IDgemini-robotics-2-2026