Google recently added interactive 3D models and simulation capabilities to its Gemini AI chatbot, allowing users to deepen their understanding through dynamic visualization when learning and researching scientific concepts, rather than relying solely on text and static diagrams. Users simply prompt Gemini to generate actionable 3D content and interactive diagrams.
Compared with previous answers that mainly displayed text and static illustrations, Gemini can now provide "functional simulations" to present complex topics in a dynamic form. Instead of passively looking at a fixed picture, users can use it to simulate complex physical systems, rotate molecular structures, or observe the moon's movement around the Earth in three dimensions. In the lunar orbit example, users can also use manual sliders to adjust parameters such as initial velocity and gravity intensity, and observe in real time how different variables work together to form a stable orbit.
To use this new feature, Google recommends users start their questions with "Show me" or "Help me visualize" and make sure to select the "Pro" model in Gemini to invoke the relevant capabilities. For example, when you enter "Visualize how fractals work", Gemini will generate dynamic images of fractal growth. Users can adjust parameters such as branch angles, length ratios, and number of iterations to observe changes in fractal morphology.
Similarly, in the field of quantum physics, users can explore the double slit interference process by "Show me how a double slit experiment works." Gemini will provide an interactive interface that allows users to modify parameters such as wavelength, wave speed, and slit distance, and see the waveform and interference pattern change in real time on the screen, allowing users to intuitively understand originally abstract physical concepts.
Google said on its blog that this 3D model and simulation feature is currently being gradually launched for all users, but users with education edition and Workspace accounts are not currently supported. According to officials, this feature is an extension of Gemini’s previous learning and teaching capabilities, with the goal of further improving its performance in education and knowledge visualization.
Previous reports indicate that Gemini can already generate clickable interactive diagrams for subjects such as biology, physics, chemistry, and mathematics, helping students explore complex academic topics graphically. Users can click on different parts of the image to obtain more detailed instructions and further extend to related sub-topics, forming a progressive learning experience. The newly added 3D models and dynamic simulations are higher-dimensional visualization capabilities expanded on this existing function.