ResearchPublished All news from that day
Ongoing case: Google's Gemini models 10 stories
Google DeepMind says Gemini 4 Argon scores 87% on the ERQA embodied reasoning benchmark.
The figure and the Boston Dynamics Spot demonstration in its 13 October robotics post are Google's own.
Check our sources · 4 facts from 1 sourceGoogle DeepMind published a robotics post for Gemini 4 Argon on 13 October 2026. It says Argon scores 87% on the Embodied Reasoning (ERQA) benchmark using pure reasoning, which Google says outperforms every other frontier model.
The post shows Argon controlling a Boston Dynamics Spot quadruped to learn to step on a foot-pedal bin, after practising in a MuJoCo digital twin it built. It also shows Argon reconstructing simulation twins from robot camera feeds, egocentric human video or hand-drawn sketches. The figures and demonstrations are Google's own.
Underlined sentences link to the facts we checked. Dotted words open a short explainer.
Fact 1 of 4 · from Google DeepMind, 13 Oct 2026
OCT. 13, 2026
Your reaction
We count reactions per story and day, never who reacted. The counts help us choose what goes in the monthly issue. If you are signed in, your own page shows yours too.
We count reactions per story and day, never who reacted. They help choose what goes in the monthly issue.
Check our sources
Every sentence above is checked against this source.
1 Gemini 4 Argon for Robotics
Open the source archived copy-
Google DeepMind published "Gemini 4 Argon for Robotics" on 13 October 2026.
citeOCT. 13, 2026
-
Google DeepMind says Gemini 4 Argon scores 87% on the Embodied Reasoning (ERQA) benchmark using pure reasoning, outperforming every other frontier model.
citeUsing pure reasoning, Argon scores 87%, outperforming every other frontier model.
-
Google DeepMind says Argon controls a Boston Dynamics Spot quadruped to learn to step on a foot-pedal trash bin to throw away soda cans, and creates an articulated MuJoCo digital twin of the bin to practise before using the physical pedal.
citeArgon controls a Boston Dynamics Spot® quadruped (12-DoF legs + 6-DoF arm) to learn a new skill: stepping on a foot-pedal trash bin to throw away soda cans. Argon creates an articulated MuJoCo digital twin of a foot-pedal trash bin to practice its moves before stepping on the physical pedal
-
Google DeepMind says Argon can reconstruct physically calibrated simulation twins from robot camera feeds, egocentric human video or hand-drawn paper sketches.
citeReconstructing physically calibrated simulation twins from robot camera feeds, egocentric human video, or hand-drawn paper sketches.
Every sentence above is checked against this source. Each fact has its own address you can cite.
Source 1 · Google DeepMind · 13 Oct 2026
Gemini 4 Argon for Robotics
Google DeepMind says Gemini 4 Argon scores 87% on the Embodied Reasoning (ERQA) benchmark using pure reasoning, outperforming every other frontier model.
Using pure reasoning, Argon scores 87%, outperforming every other frontier model.
Google DeepMind says Argon controls a Boston Dynamics Spot quadruped to learn to step on a foot-pedal trash bin to throw away soda cans, and creates an articulated MuJoCo digital twin of the bin to practise before using the physical pedal.
Argon controls a Boston Dynamics Spot® quadruped (12-DoF legs + 6-DoF arm) to learn a new skill: stepping on a foot-pedal trash bin to throw away soda cans. Argon creates an articulated MuJoCo digital twin of a foot-pedal trash bin to practice its moves before stepping on the physical pedal
Show all 4 factsShow fewer facts
Google DeepMind says Argon can reconstruct physically calibrated simulation twins from robot camera feeds, egocentric human video or hand-drawn paper sketches.
Reconstructing physically calibrated simulation twins from robot camera feeds, egocentric human video, or hand-drawn paper sketches.
Topics
The morning email
On the mornings we publish: the three top stories and up to four short ones. Free.
We email you a link to confirm. An issue may include one sponsor, always labelled Sponsored · Advertisement. Our emails count opens and clicks, not who made them. Unsubscribe in one click. What we keep