Google DeepMind unveils Gemini Robotics 2 with intelligent whole-body control for humanoids

Google reveals Gemini Robotics 2.0, promising improved dexterity and safety

Robots powered by Google's Gemini AI models are now more capable. With the debut of Gemini Robotics 2, these physical bots can now accomplish more complex tasks, continuously analyze changing environments, and collaborate with other robots. This is thanks to a trio of new sub-models, one of which is publicly available for developers starting today. Videos of robots running, dancing, and backflipping have been a staple of the Internet for years, but these machines were programmed to perform these very narrow tasks. The goal of Gemini robotics is to create a generalist robot, one that can do anything a human could do. Google DeepMind scientists sometimes call this "physical AGI." Essentially, you tell a robot what to do, and it does it. With the 2.0 release, Google says its robotics AI can control an entire humanoid robot with improved dexterity, even for machines with complex humanoid hands. This starts with Gemini Robotics ER 2, an upgraded "embodied reasoning" model that DeepMind claims is a significant leap over the previous 1.6 release. It's integrated with the Gemini Live API, giving developers the opportunity to experience that supposed leap forward. Gemini Robotics ER 2 is what's known as a vision language model (VLM). It's designed to understand instructions and the world around it. The big upgrade here is that ER 2 can process live video feeds from the robot's cameras, allowing the system to track progress as the robot lumbers from one step to the next. Google notes that Gemini Robotics ER 2 can classify video frame completeness with almost 60 percent accuracy. That's still far from perfect, but it's much better than the 1.6 release or what you can get with the visual understanding of competing AI models. Finding specific moments in video feeds is also key to completing a task correctly. When you ask a robot to pour a cup of coffee, you definitely want it to know when to stop pouring. ER 2 apparently does this much better, identifying key moments with almost 90 percent accuracy. As the robots execute multi-step tasks, the embodied reasoning model allows them to understand failures in real time. The system can then attempt that single step again rather than going back to the start. For example, the robot can just readjust its hand position and motion if a ball it's trying to pick up rolls away or someone moves a container. The new embodied reasoning release is also what gives Google DeepMind's upgraded robot AI the ability to collaborate. The video demos show Apptronik's Apollo 2 and the simpler Franka F3 Duo working together on a task without getting in each other's way. While the test robots still can't match the speed or grace of a human, the video demos include plenty of real-time footage of the robots in action, and they do seem much less hesitant than they were in past tests. Understanding is only the first step -- getting the robot to move around in the physical world is the purview of another model. After mapping out the task, the vision language model hands things over to an upgraded vision-language-action known simply as Gemini Robotics 2. This AI model generates robot actions from those instructions in the same way other generative systems create text or images. There's also a low-latency offline version of this called Gemini Robotics On-Device 2. These models are currently limited to a small group of testers. Google says the new action models are much more accurate and efficient. Even the smaller on-device version can adapt to new robot designs with just a few hours of movement data, or around 200 examples. Heading off the robot apocalypse The issues with AI hallucinations are well known at this point, but the potential harm from mistakes when the AI has a physical embodiment sharing space with humans could be much greater. With each release of Gemini Robotics, Google DeepMind has stressed that it takes this risk seriously. According to DeepMind, each layer in Gemini Robotics includes "traditional physical safety measures with robust AI safety frameworks." With the release of Gemini Robotics 2, there's a new safety benchmark called ASIMOV-Agentic. This test evaluates models across a variety of safety factors. It can assess whether an embodied reasoning agent will refuse unsafe tool calls from a VLA. It can also determine whether a given task is possible to complete safely, as well as whether the model is able to call for human assistance if it's unsure about safety. Google DeepMind notes that Gemini Robotics ER 2 is the team's safest model yet, showing robust ability to understand safety and halt actions when a human is too close to the robot. Google's new safety benchmark is available in its entirety on Hugging Face.

Google's Gemini Can Now Stomp Around as a Humanoid Robot

Google DeepMind just released a new version of its artificial intelligence model Gemini, and it can control a range of different robots -- including humanoids capable of dextrous tasks like screwing in lightbulbs and tying trash bags. Gemini Robotics 2 combines several different AI models into a single system. Taken together, they allow a robot to make sense of its surroundings and how to act in it. A vision language model (VLM), which understands images and video, can communicate with humans and reason how to perform different tasks. Two vision language action (VLA) models, trained to understand how to move in physical space, control the robot's full-body movement as well as the movements of grippers or hands. In video demonstrations shared ahead of the release, the company showed several different robots performing complex tasks autonomously using the amalgamated model. In one demo, Apptronik's Apollo 2 robot used hands from a company called Sharpa to tidy shelves. Google DeepMind trained the model to perform these tasks using a mix of human teleoperation, video examples, and simulations -- it's not yet possible for AI models to perform a wide range of complex tasks without specific training. Although Anthropic and OpenAI have taken a lead with chatbots and AI coding tools, Google has a stronger track record in robotics research, and has published important work on using AI to train robots to do useful things. The release is another sign that the search giant is betting AI will need to break free from the digital realm to realize its full potential. (It previously partnered with Boston Dynamics, a leader in legged robots, to provide the brains for those machines.) "It's another milestone in our path towards really getting towards what we call like physical AGI, which means we get a robot to do anything that a human can," Carolina Parada, head of robotics at Google DeepMind, tells WIRED. Giving frontier AI models access to robots so that they can wander around workplaces or homes and manipulate objects does, however, come with risks. Previous research has shown that using frontier AI to control robots can produce unexpected and sometimes dangerous behavior. And the idea that these models can take sudden or unwanted actions in the digital realm became apparent recently, when an unreleased AI agent developed by OpenAI hacked several systems. "The safety question is even more pressing because you're putting them in a lot of other situations," Parada says. "There's a lot of uncertainty that will show up, and so you want to be able to understand the safety question more deeply." Parada says Google takes a multi-layered approach to safety, with guardrails applied on each model layer. It's also introducing ASIMOV-Agentic, a new benchmark for measuring the safety of various AI systems collaborating to control a robot. The benchmark detects whether a command will result in harmful or uncertain outcome. The company's CEO, Demis Hassabis, previously told WIRED that he hopes to develop an AI operating system for many different robots similar to the Android operating system for smartphones.

Google DeepMind's new AI model can control a robot's entire body

Google DeepMind says the latest version of its Gemini Robotics AI model can "control entire humanoid robots." While the previous model focused on controlling a humanoid robot's upper body, Gemini Robotics 2 now supports "whole-body motions" ranging from its feet to fingertips, according to an announcement on Thursday. The new model will allow humanoid robots to perform a wider range of actions, as it allows them to walk, crouch, stretch, and manipulate objects. Videos shared by Google show how Apptronik's Apollo 2 robot can bend over to pick up a watering can, as well as find and take specific items off a shelf. Though Google DeepMind notes that its robots "have more to advance in movement speed," it adds that this update "is an important step towards the skills needed to complete more complex, real-world tasks that require whole-body coordination." Additionally, Gemini Robotics 2 supports better dexterity, as it can now control more complex, five-fingered hands. That enables robots to perform tasks like sealing a Ziploc, tying a trash bag, or unscrewing a lightbulb. Google DeepMind is updating Gemini Robotics ER (embodied reasoning) as well, a vision-language model that helps robots to analyze their surroundings, process instructions, and perform multi-step tasks. Gemini Robotics ER 2 is better at completing tasks over an extended period of time and "now understands when tasks begin and end." Google DeepMind says this update also allows multiple robots of different types to work together and complete tasks, with one video showing how Apollo 2 instructs Google's dual-arm robot to put tools inside a bin while cleaning the garage. The company notes Gemini Robotics ER 2 is its "safest robotics model to date," as it can "better detect when humans are nearby, trigger safety tool calls and bring the robot to a safe stop if someone approaches too closely." Meanwhile, Google DeepMind has brought improvements to its Gemini Robotics On-Device Model, which can run locally on a robot without an internet connection. This model can now adapt to new embodiments faster, including those with "drastically different shapes, sensors and degrees of freedom."

Google's new Gemini Robotics 2 platform allows for 'intelligent whole-body control' - Engadget

The company made a video of the bots in action doing stuff like cleaning trash and picking up watering cans. Google DeepMind just announced a new version of its artificial intelligence model Gemini that can control a range of robots. This includes humanoid robots capable of tasks like cleaning up trash and picking up watering cans. The company released a video of the Gemini Robotics 2 platform in action and the results are fairly impressive. It shows robots putting a tape into a boombox, screwing in a lightbulb and even tying up a garbage bag. The original Gemini Robotics system could already power bots that performed finicky actions like closing storage bags and folding origami, but that was done via robotic arms and hands. This update brings "intelligent whole-body control" into the mix. I'm fairly skeptical of this stuff, but Google promises the aforementioned video features "real-time footage" of "fully autonomous" robots. This operates in stark contrast to Elon Musk's Optimus robots, as a high-profile demo was proven to have used teleoperators to control the bots. Google is doing this by stuffing the robots with several different AI models, which results in a multimodal understanding of their surroundings. There's a vision language model to help with understanding and two vision language action models that control full-body and hand movement. It's worth noting that these aren't robots aren't general-purpose devices that can do just about anything. The model was specifically trained to perform every task in the video, using a mix of human teleoperation, video examples and simulations. Google DeepMind says "it's another milestone in our path towards really getting towards what we call like physical AGI, which means we get a robot to do anything that a human can," according to a statement published by Wired. There is one final caveat here. These are robots filled with AI and, well, AI gets stuff wrong. That's frustrating when looking for an answer to a query, but potentially deadly when dealing with a heavy human-sized robot. Google says it's taking a multi-layered approach to ensure future robots are safe. Each model layer has guardrails and it has introduced a new benchmark called ASIMOV-Agentic for detecting whether a command will lead to a harmful outcome. "The safety question is even more pressing because you're putting them in a lot of other situations," Carolina Parada, head of robotics at Google DeepMind, told Wired. "There's a lot of uncertainty that will show up, and so you want to be able to understand the safety question more deeply." Finally, this is all early stages. Google isn't about to roll out consumer-facing robots anytime soon. Robotics rival Elon Musk, on the other hand, promises Tesla's upcoming Optimus robots will be "the biggest product ever" that will result in $10 trillion in sales. He once stated we'd have 50,000 to 100,000 Optimus robots by 2026 and 1,000 of the little bots roaming around manufacturing facilities by the end of 2025. The latter didn't happen and the former won't happen.

Google DeepMind's Gemini Robotics 2 controls whole humanoids

Google DeepMind has released Gemini Robotics 2, a family of models that can control a humanoid from its feet to its fingertips, coordinate several robots at once, and adapt to a new machine in a few hours. It is a real step toward "physical AI," though DeepMind's own figures show the robots are still slow and clumsy at the fiddliest tasks. Google DeepMind wants one AI brain to run every robot, and it just taught that brain to use its whole body. It has released Gemini Robotics 2, a family of models that can control a humanoid from its feet to its fingertips. The system can also coordinate several machines at once and adapt to a new robot body in a few hours. It is the clearest sign yet of what the industry calls "physical AI," and Wired framed it as a real step toward "physical AGI." The pitch is simple. The same kind of model that writes your emails is now learning to walk across a cluttered room and tidy it up. From arms to the whole body The headline change is control. DeepMind's earlier robot models mostly moved an upper body to do tabletop tasks. Gemini Robotics 2 drives the entire machine. It can make a humanoid walk, crouch, stretch and balance while manipulating objects in tight spaces built for humans. In a demo, Apptronik's Apollo 2 robot heard a single instruction: put the watering can in the green bin on the bottom shelf. It walked to a table, picked up the can, stepped over to the shelves, bent down and placed it. That sounds trivial. Coordinating legs, torso, arms and hands from one prompt is not. Dexterity has improved too. The model can drive Apollo's five-fingered, 22-joint hand to tie knots and seal a ziplock bag, and it can run simpler two-fingered grippers on other platforms. "Our goal is to bring AI into the physical world and then build the intelligence layer that can be used by every robot," said Carolina Parada, DeepMind's head of robotics. Three models, one system The release is really three models. Gemini Robotics 2 is the vision-language-action model that turns what a robot sees and hears into motor commands. It handles the physical doing. Gemini Robotics ER 2 is the reasoning layer, a high-level brain that plans multi-step jobs. In its developer briefing, Google showed it tracking its own progress on a live video feed. It can call tools like Google Search, and even steer a Boston Dynamics Spot robot to fetch a snack. It also lets different robots work as a team, a wheeled machine and a humanoid splitting a task. It is available now to developers through the Gemini API and Google AI Studio. The third, On-Device 2, runs locally with no internet connection. It can be fitted to an entirely new robot body with fewer than 200 examples and a few hours of training. That last point matters, because moving a learned skill from one machine to another has long been one of robotics' hardest problems. Impressive, and still slow DeepMind was unusually frank about the limits. Its own numbers show the gap. Bloomberg reported the system could unscrew a light bulb 92 per cent of the time, but fiddlier jobs lagged badly. The Chosun Daily put the trash-bag tie at 44 per cent and the ziplock seal at 40 per cent. The robots are also slow. They pause to think through moves a person makes without a second thought. Kanishka Rao, a DeepMind robotics director, said true dexterity remains a distant goal, and that robots still learn far less efficiently than humans, who adjust after one or two mistakes. This is a pattern across the field. Rival efforts from robotics foundation-model startups, and dexterity work on rival humanoids, keep hitting the same wall. The demos dazzle, but the machines are still far from ready for the home. Safety moves up the stack As robots gain the ability to move around people, DeepMind put more weight on safety. Gemini Robotics ER 2 is its safest model yet, it says, better at spotting when a person is close and halting until the area is clear. It resumes only when the space is empty again. The company also released a benchmark, ASIMOV-Agentic, that tests whether the reasoning model will refuse an unsafe command from the action model, and whether it flags when a task is impossible or asks a human for help. Naming a robot-safety benchmark after Isaac Asimov is on the nose. The underlying worry, machines acting on flawed instructions, is not. A hardware problem Google cannot solve There is an awkward backdrop. Google makes the software, not the robots, and the hardware supply is getting political. As Axios noted, the US just moved to ban future sales of Chinese-made robots on security grounds. Many of the bodies this software might run on are built in China. Google is working with Western partners including Apptronik, Boston Dynamics and Agile Robots, and more than 100 trusted testers. Its rivals are circling the same prize. OpenAI and Nvidia are both building robot models, and everyone is chasing the same idea: one model, any body. DeepMind is careful to call this a milestone, not the finish line. Gemini Robotics 2 makes robots more general and more useful than before. It also makes plain how far a machine still is from the easy competence of a human tidying a room.

Google's Gemini Robotics 2 gives humanoid robots full-body control

Google has introduced Gemini Robotics 2, a new robotics model designed to control entire humanoid robots, giving them the ability to walk, bend, balance, and manipulate objects while completing complex tasks. The company says the system can also coordinate multiple robots and adapt to different robot bodies with only a few hours of training. The release builds on Gemini Robotics introduced earlier this year, expanding capabilities beyond tabletop manipulation to full-body movement. Google is also launching two companion models: Gemini Robotics ER 2 for high-level reasoning and Gemini Robotics On-Device 2, which runs locally on robots without requiring cloud connectivity. Unlike conventional robots that are programmed for repetitive tasks, the new models are designed to reason through actions, adapt to unfamiliar environments, and carry out multi-step jobs with minimal retraining. Google said the vision-language-action model can control humanoid robots as well as dual-arm robotic systems, enabling them to carry out household and industrial tasks using both hands or robotic grippers. One of the biggest changes is whole-body control. While earlier models focused mainly on upper-body movements, Gemini Robotics 2 enables humanoids to walk across rooms, crouch, stretch and interact with objects. During demonstrations, Google's model controlled Apptronik's Apollo 2 humanoid robot, directing it to pick up a watering can, walk across a room and place it on a lower shelf. The same model checkpoint was also used on other robot platforms, including Franka Duo, highlighting its ability to work across different robotic embodiments. Google said the model also improves dexterity. It can operate Apollo 2's five-fingered robotic hands to perform tasks such as tying trash bags, sealing Ziploc bags, and unscrewing light bulbs. On industrial platforms using two-finger grippers, it can perform precise insertion tasks, tool handling, and object placement. The company is also making its embodied reasoning model, Gemini Robotics ER 2, available through Google AI Studio and in private preview on Gemini Enterprise Agent Platform. The vision-language-action and on-device models are being released to early-access partners. Google says Gemini Robotics ER 2 serves as the robot's high-level reasoning engine, allowing it to understand spoken instructions, break them into multiple steps, and monitor progress until a task is complete. The updated model can execute task sequences lasting several minutes while making hundreds of decisions. It can also recover from failed actions and adapt to changing environments instead of following fixed routines. Another new capability is multi-robot collaboration. Different types of robots can now communicate and divide work to complete jobs that would be difficult for a single machine to perform alone. For applications where internet connectivity is unavailable, Google introduced Gemini Robotics On-Device 2. The company said the model runs locally on robotic hardware and can adapt to entirely new dual-arm robot designs with fewer than 200 training examples collected over just a few hours. Google also introduced ASIMOV-Agentic, a new benchmark designed to evaluate robotic safety, including whether robots refuse unsafe actions, recognize uncertainty, and request human assistance when needed. The company said Gemini Robotics ER 2 also improves human proximity detection, allowing robots to stop safely when people move too close.

Google updates its AI software for robots

Why it matters: It's part of a growing trend toward using AI models to make humanoid and other robots more versatile. Driving the news: The effort includes both models that control robots' individual physical actions and those designed to handle higher-level task planning. * The update builds on the push Google announced last year to bring its Gemini models to robots. What they're saying: "Gemini Robotics 2 enables robots to reason through every movement, unlocking a broad range of tasks," Google said in a blog post. * "For example, it can enable a humanoid to walk, crouch, stretch, and manipulate objects to clean up a cluttered room. It can even team up with other robots to finish the job faster." The big picture: Google's move comes as US companies may have a tougher time getting their hands on the robots needed to run such software.

Gemini Robotics 2 brings whole body intelligence to robots

From feet to fingertips -- we are teaching robots intelligent whole-body control, fine dexterity, and teamwork to complete a broad range of complex tasks For decades, we've dreamed of robots that can seamlessly step into our world and lend a hand. Now, that vision takes a significant stride forward. Most robots are pre-programmed or teleoperated for narrow, repetitive task sequences. They lack the ability to truly learn for themselves or adapt to unpredictable environments. Moreover, transferring learned skills from one robot body to another remains incredibly difficult. To take on the hardest problems at scale, robots of every shape and size need AI models giving them the ability to think, act, and interact intelligently to safely complete tasks. We demonstrated how Gemini's multimodal understanding could drive real-world action with Gemini Robotics. Today, we are introducing Gemini Robotics 2 - the intelligence layer powering the next generation of truly adaptable robots. As it takes its first literal steps, this major advance unlocks intelligent whole-body control, advanced dexterity, and multi-robot collaboration. Gemini Robotics 2 enables robots to reason through every movement, unlocking a broad range of tasks. For example, it can enable a humanoid to walk, crouch, stretch, and manipulate objects to clean up a cluttered room. It can even team up with other robots to finish the job faster. And this profound intelligence can also run locally on-device while seamlessly adapting to entirely new robotic bodies in just a few hours. We are making this possible through three highly capable models:

Google's new AI gives robots better balance, smarter hands, and teamwork capabilities

Google just made robots a lot less clumsy and a lot more helpful. Google DeepMind just brought sci-fi robots closer to reality, and it is very exciting. Building on its original Gemini Robotics model, the company has introduced Gemini Robotics 2, an AI system designed to help robots think, move, and even work together. Unlike most robots today, which rely on pre-programmed routines or someone controlling them remotely, this new system lets robots figure things out on their own when situations get unpredictable. Can robots finally handle whole-body tasks? Until now, Google's robotics models mostly handled tabletop tasks using a robot's upper body. Gemini Robotics 2 changes that by controlling the entire humanoid, from its feet all the way to its fingertips. In one demo, Apptronik's Apollo 2 robot was asked to place a watering can into a bin on a bottom shelf. The robot walked over, picked up the can, crossed the room, and set it down exactly where it belonged. The model also brings noticeably better dexterity. It can control a five-fingered robotic hand well enough to tie a knot, seal a zip-lock bag, or unscrew a bulb, and it works just as smoothly with simple two-fingered grippers for tasks like packing and sorting. Alongside this, Google is introducing a reasoning model called Gemini Robotics ER 2, which basically acts as the robot's project manager. It breaks down instructions, keeps track of multi-minute tasks, and can even get multiple robots to coordinate on the same job. There's also an on-device version built for robots without internet access, which can adapt to a brand-new robot body in just a few hours using as few as 200 examples. What makes these robots safer to work with? Safety got real attention this time around. Google introduced a new benchmark, ASIMOV-Agentic, to test whether robots know when to refuse a risky action or ask a human for help instead. Gemini Robotics ER 2 can also sense when someone gets too close and bring the robot to a safe stop. Gemini Robotics ER 2 is already live on Google AI Studio, while the rest of the models are currently rolling out to early access partners.

Introducing Gemini Robotics ER 2

Summaries were generated by Google AI. Generative AI is experimental. For robots to assist humans in everyday environments, accurate spatial reasoning is not enough. Robots must also think fast, timing their decisions and reasoning with the real-time speed of the physical world. That's why today we're launching Gemini Robotics ER 2, our most capable "embodied reasoning" model for robotics. Think of Gemini Robotics ER 2 as a high-level brain for robots. It allows robots to chat with humans, understand the physical world, and plan multi-step tasks. It then hands off motor execution to any given lower level vision-language-action (VLA) model. Gemini Robotics ER 2 can also natively call tools like Google Search to find information, or any other user-defined function. The design of Gemini Robotics ER 2 allows the robot to "think" about what comes next while simultaneously performing its actions. Gemini Robotics ER 2 represents a significant upgrade over Gemini Robotics ER 1.6. By watching continuous video feeds, robots can now track their own progress, adapt if something goes wrong, and know exactly when to move on to the next step. 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. Gemini Robotics ER 2 is now publicly available to developers via the Gemini API, Google AI Studio, and in private preview on Gemini Enterprise Agent Platform. To help you get started, we're sharing examples of how to configure the model and prompt it to power more useful physical AI tasks. Advancing physical agentic capabilities Most tasks in the physical world are complex and require multiple steps to complete. Gemini Robotics ER 2 is a physical agent, orchestrating steps for the robot and enabling it to self-correct, and generalize to more novel situations. To build an agentic setup, developers can declare low-level control interfaces -- like Vision-Language-Action (VLA) models or navigation APIs -- as tools, and stream multimodal video, audio, or text directly into the model. Gemini Robotics ER 2 improves this tool orchestration workflow. We can evaluate its performance with robots in simulation, using real-world robot control, and even pair it with a human controlling the robot remotely.

Google DeepMind debuts Gemini Robotics 2 model series for humanoid robots

Alphabet Inc.'s artificial intelligence research lab today debuted a family of models optimized to power humanoid robots. Google DeepMind says that the Gemini Robotics 2 series enables multiple autonomous machines to collaborate on a task. According to the company, it can automate chores that comprise hundreds of steps. Many humanoid robots feature a so-called dual-system AI architecture. That means they use two AI models to carry out work. The first model, which is known as an embodied reasoning algorithm, crafts a high-level plan for how to perform a task. It then sends the plan to a so-called VLA model, which turns the instructions into low-level commands for the host robot's motors. The main highlight of the Gemini Robotics 2 series is an embodied reasoning algorithm called Gemini Robotics ER 2. It enables users to describe the task that a humanoid robot should perform in natural language. According to Google, ER 2 supports tasks that comprise hundreds of steps and take several minutes to complete. The model can split a lengthy chore among several different robots to speed it up. Moreover, a tool calling feature enables ER 2 to access external cloud services. For example, it could use Google Search to clarify parts of a prompt that it doesn't understand. Humanoid robots require the ability to redo tasks that they don't complete successfully on the first try. According to DeepMind, its engineers equipped ER 2 with two features that streamline the workflow. The first feature enables the model to track the progress of a task using footage from the host robot's cameras. If the robot makes a mistake, ER 2 can identify the last step that the machine completed correctly and pick up where it left off. That removes the need to redo chores from scratch, which saves time. The other new feature makes ER 2 better than its predecessor at determining when a task is complete. The faster a humanoid robot's AI can tick off a task, the sooner it can move on to the next one. The capability also eases certain related tasks such as identifying ways to correct mistakes. "By watching continuous video feeds, robots can now track their own progress, adapt if something goes wrong, and know exactly when to move on to the next step," Google engineers Steven Hansen and Peng Xu wrote in a blog post today. After ER 2 generates a plan for how to carry out a task, it can send the instructions to one of the two other models in the Gemini Robotics 2 series. They're VLA algorithms capable of translating action plans into low-level instructions for the host robot. The first model is known as Gemini Robotics 2. Unlike certain earlier models, it can control all of a humanoid robot's components and not just its hands. As a result, the algorithm can optimize the host machine's center of gravity in a way that minimizes the risk of falls. Furthermore, Gemini Robotics 2 supports a broader range of robotic hands than DeepMind's earlier software. The second VLA model that debuted today is called Gemini Robotics On-Device 2. As the name indicates, it's designed to run directly on humanoid robots' onboard computers. DeepMind says that the model can be adapted to a new robot with a few hours of training. Developers can access ER 2 via Google Cloud, the Gemini API and Google AI Studio. The company is rolling out the model alongside a new embodied AI safety benchmark. The ASIMOV-Agentic Benchmark, as it's called, is designed to evaluate human robots' ability to avoid collisions and other risks.

Gemini Robotics 2 gives robots full-body AI control

Google DeepMind announced Gemini Robotics 2, a new version of its Gemini AI model that can control a range of robots, including humanoid machines with what the company called "intelligent whole-body control." Google released a video showing the robots cleaning trash, picking up watering cans, inserting a tape into a boombox, screwing in a lightbulb and tying a garbage bag. The earlier Gemini Robotics system powered robots performing more limited manipulation tasks such as closing storage bags and folding origami, using robotic arms and hands rather than full-body movement. Google said the new video shows "real-time footage" of "fully autonomous" robots. The source article contrasted that claim with a past Optimus demonstration by Tesla that it said had been shown to use teleoperators. The system uses several AI models to give robots multimodal understanding of their surroundings. It includes a vision language model for understanding and two vision language action models that control full-body and hand movements. The robots shown are not general-purpose machines, according to the source article. The model was trained specifically for each task in the video using human teleoperation, video examples and simulations. Google DeepMind said the work is "another milestone in our path towards really getting towards what we call like physical AGI, which means we get a robot to do anything that a human can," according to a statement published by Wired. Google said it is applying a multi-layered safety approach to the robots. Each model layer has guardrails, and the company introduced a benchmark called ASIMOV-Agentic to detect whether a command could lead to a harmful outcome. "The safety question is even more pressing because you're putting them in a lot of other situations," Carolina Parada, head of robotics at Google DeepMind, told Wired. "There's a lot of uncertainty that will show up, and so you want to be able to understand the safety question more deeply." Google said the technology remains in early stages and that it does not plan to roll out consumer-facing robots soon. The source article said Elon Musk has claimed Tesla's Optimus robots could generate $10 trillion in sales and that Tesla would have 50,000 to 100,000 Optimus robots by 2026, after a separate projection of 1,000 robots in factories by the end of 2025 did not happen.

Google unveils Gemini Robotics 2 family, expands AI models for humanoid robots

Google DeepMind unveiled Gemini Robotics 2, a new AI model family for robots. These models enable robots to move their entire bodies and handle delicate tasks. The ER 2 model plans and manages complex tasks, while On-Device 2 operates offline. Gemini Robotics 2 allows full humanoid robots to interact with objects using both hands. The company also released a safety benchmark for testing robot behavior around people. Google DeepMind on Thursday introduced Gemini Robotics 2, a new family of artificial intelligence models designed to power robots that can move their entire bodies, handle delicate tasks and work together with other robots. The announcement was made by Google AI on X, while Google DeepMind said the models are aimed at making robots more adaptable to real-world environments. The company launched three models, Gemini Robotics 2, a

Google Unveils Gemini Robotics 2 with Whole-Body Control

Google has introduced Gemini Robotics 2, a system that merges advanced artificial intelligence with robotics to address complex tasks in dynamic environments. According to AI Grid, the system features capabilities such as whole-body control and dexterous manipulation, allowing it to perform actions like maintaining balance on uneven surfaces or carefully handling fragile items in industrial settings. These features aim to enhance precision and adaptability in real-world applications. Explore how Gemini Robotics 2 facilitates multi-robot collaboration, allowing coordinated efforts in scenarios like warehouse logistics or disaster response. Learn about its generalist robotics model, which supports adaptability across tasks ranging from household assistance to hazardous work environments. Gain insight into the practical applications and broader implications of this advanced system. Whole-Body Control: Stability and Precision in Motion Gemini Robotics 2 employs whole-body control to synchronize movements across all joints and actuators, making sure stability and precision. This capability allows the robots to maintain balance, recover from disruptions and execute tasks requiring coordinated motion. * They can navigate uneven or unstable terrain while maintaining stability. * They can reach for objects or perform actions without compromising balance. This level of control is essential for tasks that demand both mobility and precision, such as maneuvering through crowded spaces or handling fragile items. By adapting to dynamic environments, these robots ensure efficiency and reliability in real-world applications. Dexterous Manipulation: Precision at Its Finest One of the standout features of Gemini Robotics 2 is its advanced hand dexterity, which enables the robots to perform intricate tasks with remarkable precision. These tasks include: * Tying knots, assembling small components, or operating delicate machinery. * Packing fragile objects with care to prevent damage. By mimicking the fine motor skills of human hands, these robots achieve exceptional accuracy in handling complex items. This capability is particularly valuable in industries such as manufacturing, healthcare and logistics, where precision and reliability are critical to success. Advance your skills in robotics by reading more of our detailed content. Multi-Robot Collaboration: Teamwork Redefined Gemini Robotics 2 introduces a innovative approach to multi-robot collaboration, allowing robots to work together seamlessly while maintaining independent functionality. Through shared reasoning and communication, these robots can coordinate their efforts effectively. For example: * In a warehouse, one robot might organize tools while another safely handles hazardous materials. * In disaster response scenarios, multiple robots could divide tasks to ensure both efficiency and safety. This collaborative capability significantly expands the range of tasks robots can perform, making them indispensable in environments that demand teamwork and adaptability. By working together, these robots can achieve outcomes that would be challenging or impossible for a single robot to accomplish. Generalist Robotics Model: Adapting to Any Scenario Unlike robots designed for specific, specialized tasks, Gemini Robotics 2 is built on a generalist robotics model, allowing it to adapt to a variety of scenarios and environments. Key features of this model include: * Recognizing and recovering from task failures to ensure consistent performance. * Autonomously retrying tasks and adjusting strategies to accommodate changing conditions. This adaptability makes Gemini Robotics 2 suitable for a wide range of applications, from assisting with household chores to performing complex industrial operations. The robots' ability to adjust to diverse situations ensures their practicality and effectiveness in real-world settings. AI Integration: Bridging Vision, Language and Action At the core of Gemini Robotics 2 is its embodied reasoning model, which integrates artificial intelligence to seamlessly connect vision, language and action. This advanced AI system enables the robots to: * Understand their surroundings and identify objects with precision. * Execute tasks based on verbal commands, such as "pick up the red cup on the table." By combining these capabilities, the robots can interact intuitively with humans and their environments. This integration ensures that Gemini Robotics 2 is not only highly functional but also user-friendly, making it accessible for both personal and professional use. Applications and Future Potential The potential applications of Gemini Robotics 2 are vast, spanning multiple industries and use cases. These robots could: * Assist with household tasks, such as cleaning, organizing, or preparing meals. * Take on hazardous industrial tasks, including handling toxic substances or operating in extreme conditions. * Support healthcare by assisting in surgeries, patient care, or rehabilitation exercises. * Enhance education and public services by navigating human-designed environments and providing interactive support. The ultimate goal is to create robots that seamlessly integrate into the physical world, enhancing human capabilities and improving quality of life. By addressing practical challenges and offering innovative solutions, Gemini Robotics 2 represents a significant step toward a future where robots and humans collaborate effectively in meaningful ways. Media Credit: TheAIGRID Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, Geeky Gadgets may earn an affiliate commission. Learn about our Disclosure Policy.

Google unveils Gemini Robotics ER 2 with video understanding and multi-robot collaboration

Google has introduced Gemini Robotics ER 2, an embodied reasoning model for robotics that combines video understanding, task orchestration and multi-robot collaboration. The model enables robots to understand the physical world, interact with humans, plan multi-step tasks and adapt their actions based on real-time situations. Gemini Robotics ER 2 works as a high-level reasoning layer for robots, while lower-level Vision-Language-Action (VLA) models handle motor execution. The model can also call external tools, including Google Search and developer-defined functions, allowing robots to gather information and complete tasks using connected capabilities. Compared with Gemini Robotics ER 1.6, the model adds improved video understanding, progress tracking and multi-robot collaboration. By analyzing continuous video feeds, robots can monitor their progress, recover from errors and determine when to move to the next step. The model can also reason about upcoming actions while a robot is performing the current task. Gemini Robotics ER 2 capabilities Gemini Robotics ER 2 acts as a physical agent that can orchestrate multiple steps, self-correct during tasks and adapt to unfamiliar situations. Developers can connect low-level control interfaces, including Vision-Language-Action (VLA) models and navigation APIs, as tools and stream multimodal video, audio and text inputs directly into the model. The model supports: * Natural language interaction with users * Multi-step task planning and orchestration * Real-time task progress tracking * Self-correction and recovery from failed actions * Tool calling through Google Search and developer-defined functions * Coordination between multiple robots Google evaluated Gemini Robotics ER 2 using simulated robots, real-world robot control and human tele-operation setups. The company said the model improves tool orchestration performance compared with Gemini Robotics ER 1.6 across real VLA, simulated VLA and human tele-operation control modes. Gemini Robotics ER 2 integrates with the Gemini Live API through a bidirectional streaming endpoint designed for latency-sensitive robotics tasks. This allows the model to coordinate action models and robotics APIs while reducing delays between reasoning and execution. Google demonstrated the model with Boston Dynamics' Spot robot, where Gemini Robotics ER 2 controls Spot APIs, including navigation and manipulator movement, to fetch objects using natural language commands. Google has also shared example code on GitHub to help developers configure and use the model for robotics applications. Video understanding and task progress tracking Gemini Robotics ER 2 improves video understanding and progress tracking, allowing robots to verify whether complex tasks, such as tightening a light bulb or tying a trash bag, are completed before moving to the next step. The update focuses on two areas: progress classification and moment-finding. Progress classification allows robots to track how far they have progressed toward completing a task. Gemini Robotics ER 2 evaluates each video frame and assigns it to one of five progress stages: * 0-20% * 20-40% * 40-60% * 60-80% * 80-100% By measuring task progress, the model provides real-time awareness and allows robots to adjust actions or retry failed steps without restarting the entire workflow. Moment-finding enables the model to identify the exact video frame where an important event occurs, such as knowing when to stop pouring coffee into a cup. Gemini Robotics ER 2 can use this capability to switch between tasks, verify completion and suggest corrections when required. The model achieves: * 91.3% accuracy on moment-finding tasks * 0.96-second mean absolute distance * 4× faster execution compared with larger model categories Multi-robot collaboration Gemini Robotics ER 2 introduces multi-robot collaboration, allowing different robots to communicate through a shared semantic understanding and coordinate tasks. Different robots can contribute based on their capabilities, such as wheeled robots operating indoors and humanoid robots handling uneven terrain. Google demonstrated this capability with Apptronik's Apollo 2 humanoid robot and Franka F3 Duo, showing how multiple robotic systems can collaborate on shared tasks. Improving spatial reasoning Gemini Robotics ER 2 improves spatial reasoning across three areas: * Success and failure detection: The model now works with raw video feeds instead of static images to identify failures during task execution, including spills, slips and misalignments. * General instrument reading: The model expands support beyond circular dials and sight glasses to digital displays, linear scales, rulers and liquid thermometers. Google tested this capability across 10 different instrument types. * Enhanced spatial Visual Question Answering (VQA): The model improves visual question answering through Gemini's multimodal understanding capabilities. Safety improvements Gemini Robotics ER 2 improves capabilities related to Safety Instruction Following and Human Proximity benchmarks. These evaluations measure how models follow physical constraints during reasoning tasks and detect nearby humans. Google said Gemini Robotics ER 2 can stop a humanoid robot when a person is nearby and resume operation only after the area is clear. The company is also introducing a benchmark to evaluate robotics foundation models as safe VLA orchestrators. The benchmark measures their ability to: * Enforce safety constraints * Monitor the environment * Assess physical feasibility * Request human clarification when needed Availability Gemini Robotics ER 2 is available to developers through the Gemini API and Google AI Studio. It is also available in private preview on the Gemini Enterprise Agent Platform.

Google Launches Gemini Robotics 2 for Humanoid Robots

Google has introduced Gemini Robotics 2, a new AI platform for humanoid robots that helps them understand, move, and complete real-world tasks across different industries. Google has taken another big step in robotics with the launch of Gemini Robotics 2. DeepMind has revealed this new AI model designed to power the next generation of humanoid robots. Tech enthusiasts can consider it a brain for humanoid robots, as it will give them the ability to reason, move, and complete complex tasks in the physical world. In the , the company has mentioned, "Think of Gemini Robotics ER 2 as a high-level brain for robots. It allows robots to chat with humans, understand the physical world, and plan multi-step tasks. It then hands off motor execution to any given lower-level vision-language-action (VLA) model." According to reports, it can lift an object off a shelf and land it back accurately around 76.3% of the time. Similarly, it can seal a zip bag properly nearly 40% of the time. When users buy, they will get three models together. First comes Gemini Robotics ER 2, which listens to instructions and breaks them into layers. The second part is Gemini Robotics 2, which turns those steps into motor commands across the whole machine. Finally, it will be Gemini Robotics On-Device 2, which runs the same work on the robot itself. Also Read: The platform is meant for many kinds of robots, not just one model. Developers can build their own hardware while using it. These ideas are still developing, and many challenges remain. Cost, safety, and reliability will decide how quickly robots become common. Even so, Google's latest platform shows that companies are preparing for a future where humans and robots work side by side instead of replacing one another.

Google Introduces Gemini Robotics Er 2

Google introduced Gemini Robotics ER 2 represents a step change in powering robots with video understanding, task orchestration, and multi-robot collaboration ? making it possible for robots to be more helpful in the physical world. For robots to assist humans in everyday environments, accurate spatial reasoning is not enough. Robots must also think fast, timing their decisions and reasoning with the real-time speed of the physical world. That?s why today we?re launching Gemini Robotics ER 2, our most capable ?embodied reasoning? model for robotics. Think of Gemini Robotics ER 2 as a high-level brain for robots. It allows robots to chat with humans, understand the physical world, and plan multi-step tasks. It then hands off motor execution to any given lower level vision-language-action (VLA) model. Gemini Robotics ER 2 can also natively call tools like Google Search to find information, or any other user-defined function. The design of Gemini Robotics ER 2 allows the robot to ?think? about what comes next while simultaneously performing its actions. Gemini Robotics ER 2 represents a significant upgrade over Gemini Robotics ER 1.6. By watching continuous video feeds, robots can now track their own progress, adapt if something goes wrong, and know exactly when to move on to the next step. 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. Gemini Robotics ER 2 is now publicly available to developers via the Gemini API, Google AI Studio, and in private preview on Gemini Enterprise Agent Platform. To help you get started, we?re sharing examples of how to configure the model and prompt it to power more useful physical AI tasks. Advancing physical agentic capabilities Most tasks in the physical world are complex and require multiple steps to complete. Gemini Robotics ER 2 is a physical agent, orchestrating steps for the robot and enabling it to self-correct, and generalize to more novel situations. To build an agentic setup, developers can declare low-level control interfaces ? like Vision-Language-Action (VLA) models or navigation APIs ? as tools, and stream multimodal video, audio, or text directly into the model. Gemini Robotics ER 2 improves this tool orchestration workflow. We can evaluate its performance with robots in simulation, using real-world robot control, and even pair it with a human controlling the robot remotely. In robotics, high-level reasoning depends on execution speed. Gemini Robotics ER 2 integrates into the Gemini Live API, using a bidirectional streaming endpoint optimized for latency-sensitive tasks. The result is fluid orchestration: Gemini Robotics ER 2 commands action models and robotics APIs to complete multi-step tasks without the jarring ?stop-and-think? pauses. To illustrate this, we?ve built a demo with Spot from our partners at Boston Dynamics. We use Gemini Robotics ER 2 to orchestrate Spot APIs, such as navigation and manipulator movement, creating an interactive robot that fetches objects for you. The code is available on Github with other examples. Unlocking temporal intelligence for robust task completion One of robotics? hardest challenges is knowing when a task is done. Gemini Robotics ER 2 brings a step-change in video understanding and progress tracking to verify that complex tasks ? such as tightening a light bulb or tying a trash bag ? are complete to specification before switching to the next task. In this update, we?ve made progress on two foundational capabilities for task progress understanding: progress classification and moment finding. Continuous progress classification Progress classification refers to a robot?s ability to track progress towards task completion. In our evaluations, we assign each frame in a video feed into five levels of progress (0-20%, 20-40%, 40-60%, 60-80%, 80-100%). By quantifying task progress, Gemini Robotics ER 2 provides robots with real-time situational awareness, and allows them to adjust actions on the fly or retry failed steps without restarting an entire workflow. Precision moment-finding Moment-finding measures a model's ability to identify the exact video frame where a critical event takes place (i.e. when to stop pouring coffee into a cup). Gemini Robotics ER 2 achieves significant gains in performance on moment finding, enabling robots to precisely switch between tasks, verify success and suggest corrections. Multi-robot collaboration No single robot fits every task ? a wheeled rover excels indoors, while a humanoid robot may excel at uneven terrain. Gemini Robotics 2 enables multi-robot collaboration, allowing diverse machines to communicate via a shared semantic understanding to handoff and complete complex tasks. See how Gemini Robotics ER 2 enables Apptronik?s Apollo 2 and Franka F3 Duo to collaborate here. Improving general spatial intelligence Gemini Robotics ER 2 advances our core spatial reasoning capability, as measured by three benchmarks: Success/failure detection: Now operates on raw video feeds rather than static snapshots to catch mid-execution failures like spills, slips, or misalignments. General instrument reading: Extends beyond circular dials and sight glasses to include digital displays, linear scales, rulers, and liquid thermometers. We tested it across 10 different types of instruments. Enhanced spatial VQA: Improves Visual Question Answering throughGemini?s advancements in multi-modal understanding. Advancing safety for embodied intelligence Gemini Robotics ER 2 is our safest model, achieving significant gains on Safety Instruction Following and Human Proximity benchmarks, which evaluate how a model adheres to physical constraints during reasoning tasks and spatial awareness for detecting humans. We found that Gemini Robotics ER 2 successfully halts a humanoid robot when a person is nearby and autonomously resumes work only once the area is clear. To advance safety for physical agents, we?re introducing a benchmark that evaluates a foundation model's ability to act as a safe VLA orchestrator by testing its capacity to enforce safety constraints, monitor the environment, assess physical feasibility, and seek human clarification. For details, see our safety technical report. Looking ahead, our plans are to push these models towards even more complex tasks to accelerate the development of helpful robots and support the robotics community.