Google DeepMind Unveils Gemini Robotics 2 for Whole-Body Control and Advanced Dexterity
Google DeepMind has unveiled Gemini Robotics 2, a foundational intelligence layer designed for next-generation physical robots. This model bridges multimodal understanding and real-world actions, seamlessly coordinating complex whole-body movements like walking, squatting, and reaching with highly precise fingertip manipulation. The system also supports collaborative tasks, allowing groups of robots to divide work and coordinate actions autonomously as a team.
Related tools
Recommended tools for this topic
These picks prioritize high-intent tools relevant to this topic. Some links may include partner or affiliate tracking.
Strong fit for AI, backend, and frontend readers looking for an AI-first coding workflow.
View CursorNatural next step for readers evaluating LLM adoption, APIs, and production inference.
Explore APIA strong fit for readers comparing Claude-class models, safety, and long-context workflows.
View AnthropicComparison
| Aspect | Before / Alternative | After / This |
|---|---|---|
| Control scope | Pre-programmed specific movements or narrow teleoperated tasks | Integrated whole-body coordination, from navigation to finger-level manipulation |
| Environment adaptability | Fragile performance in dynamic, unpredictable, or unstructured environments | Dynamic adaptation through real-time multimodal reasoning and pathfinding |
| Skill transferability | Highly specialized routines that cannot be easily transferred to different hardware | Cross-platform capability allowing learned physical skills to run on various robotic chassis |
| Coordination | Isolated single-robot execution requiring central scripting | Autonomous multi-robot collaboration and dynamic task division |
Source: DeepMind Blog
This page summarizes the original source. Check the source for full details.
