Gemini Robotics-ER 1.6: Enhanced Embodied Reasoning for Robots
ID: gemini_robotics_er16_2026_001
Author: Google DeepMind
Source: <https://deepmind.google/technologies/gemini-robotics/>
Date: 2026-04-14
Category: models
Quality Score: 4
Tags: Gemini, robotics, embodied-reasoning, spatial, physical-reasoning, 2026
Fetched: 2026-06-07 20:30:54
One-liner: Gemini Robotics-ER 1.6将LLM推理能力延伸到物理世界，机器人理解环境的方式正在被重新定义
摘要 (zh): Google DeepMind于2026年4月14日发布gemini-robotics-er-1.6-preview，这是一个更新的机器人模型，新增了仪器读取和改进的空间与物理推理能力。该升级旨在增强机器人的具身推理能力，使它们能够更好地理解物理环境并与之交互。该模型取代了4月30日关闭的gemini-robotics-er-1.5-preview版本。
English (Original)
Gemini Robotics
原文链接: https://deepmind.google/technologies/gemini-robotics/
https://deepmind.google/models/gemini-robotics/gemini-robotics/
Gemini Robotics 1.5
Our most capable vision-language-action (VLA) model. It can ‘see’ (vision), ‘understand’ (language) and ‘act’ (action) within the physical world. It processes visual inputs and user prompts, learning within different embodiments and increasing its ability to generalize problem-solving.
[
Learn more
](https://deepmind.google/models/gemini-robotics/gemini-robotics/)
中文翻译
_等待人工或后续流水线翻译。_