Google DeepMindUnited Kingdom2026

Gemini Robotics ER 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.

RobotsSpotby Boston Dynamics BApollo 2by Apptronik B

Key figures as stated by the developer

4 of 6 stated · 2 gaps

Model type ⁠B embodied reasoning
Controls not stated
Inputs ⁠A camera language audio
Access ⁠B API only
Parameters not stated
Base model ⁠A Gemini 3.5 Flash

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 (5)

Model type

Developer wording: "our most capable “embodied reasoning” model for robotics".

Controls

Note The model returns text, which has no value in the closed list: "Text, with up to 64K token output." The movement is done by a separate vision-language-action model: "It then hands off motor execution to any given lower level vision-language-action (VLA) model."

Inputs

Model card, Inputs: "Text / Image / Video / Audio (interleaving possible)."

Access

"Gemini Robotics ER 2 is now publicly available to developers via the Gemini API". "Our models are available to downstream providers via an application program interface (API) and subject to relevant terms of use."

Base model

"Gemini Robotics ER (Embodied Reasoning) 2 is based on Gemini 3.5 Flash."

Full specifications

5 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 type	`embodied reasoning`	B Developer wording: "our most capable “embodied reasoning” model for robotics".
Inputs	`camera` `language` `audio`	A Model card, Inputs: "Text / Image / Video / Audio (interleaving possible)."
Controls	not stated	Note The model returns text, which has no value in the closed list: "Text, with up to 64K token output." The movement is done by a separate vision-language-action model: "It then hands off motor execution to any given lower level vision-language-action (VLA) model." ⁠A
Degrees of freedom controlled	not stated
Size and speed
Parameters	not stated
Control rate	not stated
Where it runs and what it is built on
Runs where	not stated	Note "There is no required hardware or software to use the model." ⁠A
Base model	Gemini 3.5 Flash	A "Gemini Robotics ER (Embodied Reasoning) 2 is based on Gemini 3.5 Flash."
Openness
Access	API only	B "Gemini Robotics ER 2 is now publicly available to developers via the Gemini API". "Our models are available to downstream providers via an application program interface (API) and subject to relevant terms of use."
License	not stated	Note "Our models are available to downstream providers via an application program interface (API) and subject to relevant terms of use." ⁠A
The developer’s own claims
Learns new tasks	not stated
Works across robots	yes	B Developer claim: "It then hands off motor execution to any given lower level vision-language-action (VLA) model." The announcement shows it orchestrating Boston Dynamics Spot and Apptronik Apollo 2 with Franka F3 Duo. "We are also introducing multi-robot collaboration, enabling robots to work together in shared spaces and complete complex workflows a single robot could not do alone." 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

Google DeepMind calls Gemini Robotics ER 2 its most capable embodied reasoning model for robotics; it plans tasks for a robot and hands the movement to a separate vision-language-action model.
Google DeepMind announced the model on 30 July 2026.

Sources

A Manufacturer product page. deepmind.google/models/model-cards/gemini-robotics-er-2/ · Retrieved 2026-10-13
B blog.google/innovation-and-ai/models-and-research/google-deepmind/gemini-robotics-er-2/ · Retrieved 2026-10-13

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.