---
schema_version: "newruntime-agent-readable-v0.2"
type: "post"
stable_id: "post:gemini-robotics-er-2-control-plane"
slug: "gemini-robotics-er-2-control-plane"
title: "Gemini Robotics ER 2 Is A Control Plane For Physical Agents"
description: "Gemini Robotics ER 2 acts as a high-level embodied reasoning model that watches video, calls tools, plans multi-step tasks, and coordinates robot collaboration."
retrieval_nugget: "Gemini Robotics ER 2 acts as a high-level embodied reasoning model that watches video, calls tools, plans multi-step tasks, and coordinates robot collaboration. #Gemini #Robotics #AgentControlPlane #VideoUnderstanding Gemini Robotics ER 2 is not a \"robot model\" in the casual sense. It is a control plane for a physical agent."
status: "published"
published_at: "2026-07-31"
updated_at: "2026-08-01"
record_date: "2026-08-01"
date_kind: "updated_at"
topics: ["robotics","agent-control-plane","video-understanding","gemini"]
source_urls: ["https://deepmind.google/blog/gemini-robotics-er-2-powering-robotics-with-video-understanding-task-orchestration-and-multi-robot-collaboration/"]
visuals: [{"id":"gemini-robotics-er-2-control-plane","kind":"editorial-diagram","role":"hero","src":"https://newruntime.com/images/posts/gemini-robotics-er-2-control-plane.webp","alt":"A whiteboard diagram showing continuous video feedback entering an embodied reasoning control plane that calls tools, delegates to action models, and coordinates multiple robots.","caption":"Gemini Robotics ER 2 is positioned as the high-level control plane: watch progress, call tools, delegate motor execution, and coordinate robots without stop-and-think pauses.","credit":"New Runtime synthesis from Google Gemini Robotics ER 2","source_url":"https://deepmind.google/blog/gemini-robotics-er-2-powering-robotics-with-video-understanding-task-orchestration-and-multi-robot-collaboration/","generated_with":"gemini-3.1-flash-image","width":1600,"height":900,"legend":[{"label":"Video feedback","description":"Continuous observation tells the agent whether a physical step is complete."},{"label":"Tool orchestration","description":"The reasoning model calls search, functions, robot APIs, and VLA controllers."},{"label":"Collaboration","description":"Multiple robots can divide work that one body cannot complete alone."}]}]
telegram_message_id: 2843
telegram_url: "https://t.me/qwgai/2843"
telegram_message_ids: [2843]
telegram_delivery_mode: "rich_media"
telegram_media_url: "https://t.me/qwgai/2843"
routes: {"html":"https://newruntime.com/posts/gemini-robotics-er-2-control-plane/","markdown":"https://newruntime.com/posts/gemini-robotics-er-2-control-plane.md","json":"https://newruntime.com/posts/gemini-robotics-er-2-control-plane.json"}
source_format: "markdown"
---

# Gemini Robotics ER 2 Is A Control Plane For Physical Agents

## Retrieval answer

Gemini Robotics ER 2 acts as a high-level embodied reasoning model that watches video, calls tools, plans multi-step tasks, and coordinates robot collaboration. #Gemini #Robotics #AgentControlPlane #VideoUnderstanding Gemini Robotics ER 2 is not a "robot model" in the casual sense. It is a control plane for a physical agent.

#Gemini #Robotics #AgentControlPlane #VideoUnderstanding

Gemini Robotics ER 2 is not a "robot model" in the casual sense. It is a control plane for a physical agent. Google describes it as an embodied reasoning model: it understands video, plans multi-step tasks, calls tools, and hands motor execution to a lower-level VLA model or robot API.

The key detail is progress understanding. In the physical world, the system cannot simply execute a step and move down a checklist. The model has to see that the light bulb is actually tightened, the bag is actually tied, and the object is actually delivered; if something goes wrong, it has to adjust the path to the next step.

Another important piece is the parallel with agent platforms: ER 2 can reason about the next action while the robot continues the current one, and it can coordinate multiple robots in one space. This is no longer a "plan -> act -> check" loop; it is a runtime with streaming perception, tools, delegation, and safety.

For software agents, the lesson is direct: orchestration wins not when the model is smarter in isolation, but when state, progress, and completion become first-class entities in the system.