本文为原文前 6,000 字符的节选翻译，完整内容请查看原文。

Gemini 4 Argon for Robotics

Gemini 4 Argon marks a step change in physical understanding, code generation, and visual-semantic reasoning—unlocking new workflows across the robotics stack and accelerating progress toward physical AGI.

Gemini 4 Argon 在物理理解、代码生成和视觉语义推理方面实现了跨越式发展，为机器人技术栈解锁了新的工作流程，并加速了通向物理通用人工智能（AGI）的进程。

You can see this leap in capability on the Embodied Reasoning (ERQA) benchmark, which tests a model’s ability to understand and solve physical, spatial problems in 3D environments. Using pure reasoning, Argon scores 87%, outperforming every other frontier model.

你可以在具身推理（ERQA）基准测试中看到这种能力的飞跃，该测试旨在评估模型在 3D 环境中理解和解决物理及空间问题的能力。凭借纯粹的推理能力，Argon 获得了 87% 的分数，超越了所有其他前沿模型。

As a general-purpose physical reasoning model, Argon can be applied across the entire robotics lifecycle. In this post, we share representative prompts for Argon across four key areas: Robot control, Real2Sim, Physical understanding, and Research and self-improvement.

作为一种通用物理推理模型，Argon 可以应用于整个机器人生命周期。在本文中，我们分享了 Argon 在四个关键领域的代表性提示词：机器人控制、实物到仿真（Real2Sim）、物理理解以及研究与自我提升。

Unlocking robot control with Argon

利用 Argon 解锁机器人控制

1. Spot mobile manipulation: Putting in the trash bin with MuJoCo practice Argon controls a Boston Dynamics Spot® quadruped to learn a new skill: stepping on a foot-pedal trash bin to throw away soda cans. Argon creates an articulated MuJoCo digital twin of a foot-pedal trash bin to practice its moves before stepping on the physical pedal and dropping soda cans into the open bin.

1. Spot 移动操作：通过 MuJoCo 练习进行垃圾投放 Argon 控制波士顿动力公司的 Spot® 四足机器人学习一项新技能：踩踏脚踏式垃圾桶以丢弃汽水罐。Argon 创建了一个脚踏式垃圾桶的 MuJoCo 铰接式数字孪生模型，在踩下物理踏板并将汽水罐投入打开的垃圾桶之前，先进行动作练习。

2. Zero-shot control Argon controls a simple robotic arm to place a fallen mug upright and put the crumpled napkin inside. Argon reasons with a visual coordinate harness and action primitives to perform this task over several turns.

2. 零样本控制 Argon 控制一个简单的机械臂将倒下的马克杯扶正，并将揉皱的餐巾纸放入其中。Argon 通过视觉坐标系和动作原语进行推理，在几个回合内完成了这项任务。

3. Omni control Argon steers robots in the Gemini Omni World Model to rearrange five scattered Scrabble tiles (G, R, A, N, O) into a clean row spelling A - R - G - O - N. Instead of specifying one static prompt, Argon steers in a closed visual loop. It writes timestamped action prompts, inspecting intermediate rollout keyframes, and refining its prompts.

3. 全能控制 Argon 在 Gemini Omni 世界模型中引导机器人，将五个散乱的拼字游戏字母块（G, R, A, N, O）重新排列成整齐的一行，拼出 A - R - G - O - N。Argon 并非使用单一的静态提示词，而是在闭环视觉中进行引导。它编写带有时间戳的动作提示词，检查中间的推演关键帧，并不断优化其提示词。