Compare

Google DeepMind · robot AI model

Gemini Robotics 2.5

United Kingdom · 2026

Runs on Apollo 2 by Apptronik

Sources retrieved: 10 November 2026 (A)

Model type
Vision-language-action (VLA)⁠
What it controls
Arms and hands⁠
Access
Partners / early access only⁠

Visit website

White humanoid robot with a rounded head and silver five-finger hands tying a ribbon around a gift box at a kitchen counter

Specifications

5 of 14 stated 9 not stated

How to read the figuresShow guideClose
  • not statedNo source
  • yesStated by a source
  • 0Reported as zero
  • ASource

Select a source marker for source details and notes.

Release1/1 disclosed

Released10 November 2026

Input and output2/4 disclosed

Model typeVision-language-action (VLA)
InputsNot stated
What it controlsArms and hands
Degrees of freedom controlledNot stated

Size and speed0/2 disclosed

ParametersNot stated
Control rateNot stated

Where it runs and what it is built on0/2 disclosed

On robot or cloudNot stated
Base modelNot stated

Openness1/3 disclosed

Licence documentNot stated
Licence typeNot stated

The developer’s own claims1/2 disclosed

Learns new tasksNot stated
Works across robotsYes

Notes

Google DeepMind introduces Gemini Robotics 2.5 on 10 November 2026 as “Gemini Robotics 2.5, our most capable vision-language-action (VLA) model to date”. It is a separate model from Gemini Robotics 2, which Google DeepMind lists as a related earlier post: “Gemini Robotics 2 brings whole body intelligence to robots”.

In a proof-of-concept on Mercedes-Benz production lines, Google DeepMind pairs the Gemini Robotics ER reasoning model with the Gemini Robotics 2.5 “actions model to control its hands and arms” on Apptronik's Apollo 2 humanoid.

Sources

One line per source, not per figure: a source has one retrieval date.

  1. Adeepmind.google/blog/gemini-robotics-2-5-accelerates-real-world-physical-intelligence/Main source. Retrieved 10 November 2026

Wrong or missing data? Tell us

More models

Same developer

Same model type