On April 10, Google Gemini announced a major new feature: the system can now not only answer questions through text but also directly generate interactive 3D models and physics simulation scenarios. This marks another step forward for AI, from a simple "content generator" to an "intuitive teaching tool."

image.png

Core Experience: From "Reading Instructions" to "Adjusting Parameters"

The core of the new feature lies in its high level of interactivity. When users ask Gemini about physical processes or three-dimensional space, the AI will present a dynamic window:

  • Multi-dimensional Perspective: Users can freely drag the model to perform 360-degree rotation, or zoom in to observe specific details.

  • Parameter Linkage: The interface comes with dedicated sliders and switches. For example, when simulating the "moon's orbit around the Earth," users can adjust the orbital speed in real time or enable the display of the orbit, intuitively experiencing the laws of celestial motion.

  • Physical Feedback: Users not only see, but also input different values to observe the changes in the physics simulation in real time, making abstract concepts concrete.

Industry Competition: Visualization Becomes the "Battleground"

Currently, top global large model vendors are accelerating their layout in "visual answers," striving to make AI answers more persuasive:

  • Anthropic: Previously introduced the function of automatically generating charts and interactive diagrams for Claude.

  • OpenAI: Added specialized visualization tools for mathematical and scientific concepts to ChatGPT.

  • Google: This time, Gemini expanded its capabilities into the 3D field, further solidifying its leading position in multimodal interaction.

Getting Started Guide: How to Start the "3D Classroom"?

Currently, all Gemini users can experience this feature by switching to the Pro model. The operation is very straightforward:

  1. Switch Model: Select the "Pro" mode in the app interface.

  2. Submit Request: Try sending commands such as "Show me a double pendulum system" or "Help me visualize the Doppler effect."

  3. Click Generate: Below the text description, click the newly appeared "Show me the visualization" (Show the visualization) button to call up the interactive 3D scene.

Conclusion: Breaking Through the "Ceiling" of Understanding

With the implementation of Gemini's ability to generate interactive 3D models, AI is evolving from a "parrot" into a "digital sandbox." For professional users in education, engineering, and research fields, this intuitive visual feedback will greatly shorten the cognitive path from "acquiring knowledge" to "understanding."