Technology
Gemini 4 Hydrogen Launches Faster Robotics Model with New Antigravity Plugin
By Alex Gruenstein, Gabe Cohen, Gemini Robotics Team

Over the past month, Gemini 4 Argon and Gemini 4 Carbon demonstrated advanced physical reasoning for robotics, including 3D scene understanding, real‑environment reconstruction in simulation, task practice in simulation before hardware deployment, and direct control of robot arms from camera images. Last week, Gemini 4 Hydrogen was introduced, offering strong spatial and embodied reasoning in a faster, more efficient model. To help developers use Hydrogen, Google is launching a Robotics plugin on Antigravity, a set of skills that streamline hardware onboarding, simulation, and robot control, turning Antigravity into a one‑stop robotics workbench. The plugin allows a developer to move from a robot in a box to a working application in an afternoon.
To install the plugin, open Antigravity, go to Customizations, and search for Robotics on the Build With Google shelf. Gemini 4 Argon and Carbon showed that a general model can drive a robot arm directly, and Gemini 4 Hydrogen preserves that ability. Give it a goal in plain language and a camera view, and it plans the task, writes and runs the control code, and keeps adjusting until the job is done.
On the Artificial Analysis Robotics 0.1 benchmark, Gemini 4 Hydrogen controls a real 6‑axis arm through eight everyday tabletop tasks, from pouring a drink to packing blocks into a box. Hydrogen scores 58, similar to Argon at a lower cost per token. Hydrogen incorporates the specialized skills of the Gemini Robotics ER 2 (Embodied Reasoning) models into the core Gemini 4 model series and beats them on nearly every axis. On the public ERQA benchmark, Hydrogen outperforms Gemini Robotics ER 2 on spatial reasoning tasks such as multi‑view correspondence and 3D trajectory reasoning.
For developers using Gemini Robotics ER 2 today, the recommendation is to try Gemini 4 Hydrogen for these workloads. A fast, capable model is only half the story. Building a robot application has always required gluing together drivers, calibration scripts, simulators, perception code, and a UI, which demands robotics expertise and time. The new Robotics plugin packages that work into skills that Gemini can use on your behalf.
An end‑to‑end workflow looks like this: a developer plugs an I2RT YAM arm and an external depth camera into a laptop and asks Hydrogen in Antigravity to bring it up. The plugin’s Skills install the software, find the camera, calibrate the arm, and run test motions. The arm is calibrated and ready for tasks in about 10 minutes. A similar flow also brings up a low‑cost SO‑ARM101 arm and an industrial‑grade FR3 Duo from scratch to live waving.
Once the robot is live, a simple prompt can create a basic task within minutes. The next step is building a trustworthy simulation. With the Robotics plugin, MuJoCo runs directly inside Antigravity with dedicated creation, auditing, and rendering skills. This allows developers to build and preview high‑quality simulated scenes and assets—whether described, imported from links/documents, or uploaded via workspace photos—right alongside their code. Developers can then adjust shapes, positions, and physical properties through chat to verify reaching and grasping before deploying to physical hardware.
The workbench with Hydrogen and the Robotics plugins lets developers go from an idea to a repeatable working application in the real world in a much shorter time by iterating and validating in simulation. For example, a developer starts with a whiteboard sketch of a conveyor belt carrying mixed snacks and a robot arm picking and sorting them into bins. From one prompt of the task with images of a real scene, Hydrogen constructs an interactive MuJoCo simulation of the cell, tests and validates the grasping policy in simulation across multiple runs, and then deploys the verified policy directly to the robot. When nuances emerge, such as tuning gripper closure force or refining sorting rules, the developer iterates in natural language, asking Hydrogen to adjust the controller. Within about an hour end‑to‑end, the arm reliably sorts snacks off the moving belt.
Hydrogen’s efficiency makes it practical to run a multi‑step robot application all day. In another example, a developer uses Hydrogen to build a small web storefront for a soap shop and connects it to two YAM Pro arms. When an order comes in, the arms set out a packing bin, pick the ordered mint and citrus bars, and pull a printed thank‑you note from a printer. With just a few prompts, the FR3 Duo simulates a chess board and can play chess against itself or take on human opponents.
Gemini 4 Argon pushes the ceiling of physical reasoning. Gemini 4 Hydrogen makes those capabilities fast and affordable enough to use every day, and the Antigravity Robotics plugin puts them in a workbench where anyone can go from a robot in a box to a working application. Developers can share their own creations and use cases on the plugin’s GitHub discussions page. If they have ideas, custom skills, or derived work that would be helpful, they can let Google know, and the plugins will continue to improve.