Google Announces Gemini Robotics, Gemini 2.0 Model Optimized for Robots

Google DeepMind has been making steady progress in the field of AI with regular, highly-regarded updates to Gemini, Imagen, Veo, Gemma, and AlphaFold. Today, Google’s AI team continues to make headlines by announcing its official entry into the robotics industry with the release of two new models based on Gemini 2.0: Gemini Robotics and Gemini Robotics-ER.

Gemini Robotics: Advanced Vision-Language-Action Model

Gemini Robotics is an advanced vision-language-action (VLA) model that builds on Gemini 2.0, adding physical actions as a new output method for controlling robots. Google claims that this new model can understand situations that it has not even encountered during training.

Compared to other leading VLA models, Gemini Robotics performs twice as well on a comprehensive set of generalization benchmarks. Because it is built on the Gemini 2.0 model, it is able to understand a wide range of natural languages, which means it can understand human commands more accurately.

In terms of dexterity, Google claims that Gemini Robotics can handle complex, multi-step tasks that require precise manipulation. For example, the model can fold origami or put snacks into Ziploc bags.

Gemini Robotics-ER: A Visual-Language Model Focusing on Spatial Reasoning

Gemini Robotics-ER is an advanced visual-linguistic model focused on spatial reasoning, allowing roboticists to integrate with their existing low-level controllers. Using this model, roboticists will have all the steps to control a robot immediately, including perception, state estimation, spatial understanding, planning, and code generation.

The Future of Gemini Robotics

Google is partnering with Apptronik to build humanoid robots based on the Gemini 2.0 models. Google is also working with a number of trusted testing partners, including Agile Robots, Agility Robotics, Boston Dynamics, and Enchanted Tools, to guide the future development of Gemini Robotics-ER.

By enabling robots to understand and perform complex tasks with greater accuracy and adaptability, Google DeepMind is paving the way for a future where robots can seamlessly integrate into many aspects of our lives.

Sign up and earn $1000 a day ⋙

Leave a Comment

Google splits with Qualcomm, opts for MediaTeks 5G modem for Pixel 10 series

Google splits with Qualcomm, opts for MediaTeks 5G modem for Pixel 10 series

Google has decided to end its long-standing partnership with Qualcomm and instead use MediaTek's T900 modem in the Pixel 10 series.

Perplexitys Social Search Needs These 3 Features to Compete with Google

Perplexitys Social Search Needs These 3 Features to Compete with Google

Perplexity’s regular search engine is great, but its Social Search feature leaves a lot to be desired. Before Perplexity can even think about competing with Google in this area, it needs these new features.

This little change will make accessing your Google passwords much easier!

This little change will make accessing your Google passwords much easier!

While Google's Password Manager is a reliable solution, to access it you have to dig through Chrome's settings.

Geminis Free Version Just Removed a Major Limitation

Geminis Free Version Just Removed a Major Limitation

As one of the most powerful text-to-image AI models, Google's Imagen 3 is already available on Gemini apps, but only to a certain extent.

New Gmail scam from... Google?

New Gmail scam from... Google?

Not every account security email you receive is legitimate. And if you see an email from Google in your Gmail inbox, think twice. There's a new Gmail scam going around — and it looks like it's coming straight from Google.

How to protect your Google account with Private Checkup

How to protect your Google account with Private Checkup

Google does a great job of trying to keep all that information as private as possible, but it can't hurt to take a look at it by knowing how to secure your Google account with the Privacy Checkup tool.

ChromeOS Just Copied One of Windows 11s Best Features

ChromeOS Just Copied One of Windows 11s Best Features

Most people would probably agree that Windows 11 is not a perfect operating system. However, it is not all bad, and in fact, Windows 11 contains a lot of useful features that many people do not know or do not take advantage of.

Googles AI Mode can now view and search images

Googles AI Mode can now view and search images

Google is adding multimodal capabilities to its search-focused AI Mode chatbot, allowing it to view and answer questions about images, while expanding access to AI Mode to millions more users.

Googles AI Can Design Chips Faster, Better Than Humans

Googles AI Can Design Chips Faster, Better Than Humans

With the help of a complex neural network architecture based on edge graphs, Google Brain's AI model can design floorplans in a fraction of the time it takes humans.

ChatGPT increases users, Google slightly decreases

ChatGPT increases users, Google slightly decreases

The Internet search market is witnessing an interesting turning point, as ChatGPT gradually becomes a formidable “emerging competitor” to the giant Google.

Gmail releases extremely useful email encryption feature

Gmail releases extremely useful email encryption feature

Gmail just turned 21, and Google chose to celebrate its special birthday by launching a very meaningful feature for users: an extremely easy and useful automatic email encryption feature.

Amazon Announces Nova Sonic Sound Model, Claims Performance Surpasses OpenAI and Google

Amazon Announces Nova Sonic Sound Model, Claims Performance Surpasses OpenAI and Google

Amazon today introduced Nova Sonic, an advanced speech-to-speech model that enables developers to build apps that can converse with human-like voices in real time.

YouTube hides countdown timer to skip ads

YouTube hides countdown timer to skip ads

YouTube has just rolled out a change to ad skipping on both desktop and mobile apps.

4 Ways to Stop Google from Showing Personalized Results

4 Ways to Stop Google from Showing Personalized Results

Google often collects personal data, search history, activity, and user location to be able to show you personalized search results.

Google may soon block sideloaded apps, dealing a blow to Android freedom

Google may soon block sideloaded apps, dealing a blow to Android freedom

In an effort to prevent apps from misusing the Accessibility API, Google is planning to introduce a new set of restrictions on sideloading apps on Android 13.

How to use Aperty to edit portrait photos

How to use Aperty to edit portrait photos

Many people use Aperty for portrait editing – it has traditional RAW development tools focused on editing portraits for great results and innovative AI tools dedicated to improving professional portraits.

How to arrange overlapping images in Word

How to arrange overlapping images in Word

Overlapping images in Word is very simple when you just need to adjust the image position in Word. The article below will guide you to overlap images in Word.

5 ways to open the Startup Repair tool on Windows

5 ways to open the Startup Repair tool on Windows

Startup Repair is a Windows recovery tool that can fix some system problems that prevent Windows from starting. Startup Repair scans your PC for problems and then attempts to fix them so your PC can start correctly.

4 reasons you need a tripod to take photos on your smartphone

4 reasons you need a tripod to take photos on your smartphone

While you can take many types of photos manually on your smartphone, there are some situations where a tripod will be needed.

How to listen to music privately on Apple Music

How to listen to music privately on Apple Music

Although Apple Music doesn't have an anonymous listening mode, we can also adjust some settings to be able to listen to music on Apple Music more privately.

How to use Gemini 1.5 Flash for free

How to use Gemini 1.5 Flash for free

At I/O 2024, Google announced a number of new AI models, upcoming projects, and a plethora of AI features that will be available across its products. However, the most notable one was the Gemini 1.5 Flash model.

How to use ChatGPT o3-mini for free

How to use ChatGPT o3-mini for free

The ChatGPT o3-mini inference model is now available for free to all ChatGPT users. OpenAI has added a Reason button next to the ChatGPT content importer to use the o3-mini model.

Scammers Are Using Deepseek to Steal User Data

Scammers Are Using Deepseek to Steal User Data

Bad guys are creating thousands of DeepSeek-like websites in the hopes that unsuspecting users will give them their personal information.

How to read five/five, four/four, one/one… correctly in the sequence of natural numbers?

How to read five/five, four/four, one/one… correctly in the sequence of natural numbers?

Five or five? Are you wondering whether reading two thousand twenty-five is correct? This article will give you the answer.

How to store avocado in the freezer

How to store avocado in the freezer

Freezing can help keep avocados fresher longer, but it can reduce their vitamin content over time. Here's how to best store avocados in the freezer.

Instructions to change the default File Explorer folder

Instructions to change the default File Explorer folder

By default, File Explorer opens to the Home folder, which contains recently used folders and files. If you want to change the default File Explorer folder, follow the instructions below.

Ways to reduce the risk of birth defects in the fetus

Ways to reduce the risk of birth defects in the fetus

Birth defects are something no one wants. Although they cannot be completely prevented, you can take the following steps to reduce the risk of birth defects in your baby.

What planets are Venus and Venus?

What planets are Venus and Venus?

What are the evening star and the morning star? Here's what you need to know about the evening star and the morning star.

Huawei Mate XT Ultimate screen repair price is as expensive as an iPhone 16 Pro Max

Huawei Mate XT Ultimate screen repair price is as expensive as an iPhone 16 Pro Max

According to the official price list announced by Huawei itself, repairing the Mate XT Ultimate screen will cost up to 7,999 CNY, equivalent to 1,123 USD or nearly 28 million VND, equal to the price of an iPhone 16 Pro Max.

Summit supercomputer is about to retire

Summit supercomputer is about to retire

Oak Ridge National Laboratory (ORNL) announced that Summit, the world's most powerful supercomputer in 2018 and 2019, will be shut down in November after nearly six years of operation.