To make a 3D model in Gemini, you primarily use Google's Nano Banana model within the Gemini AI Studio. The process involves these steps:

1. Open Google Gemini AI Studio (via app or web).
2. Select the Gemini 2.5 Flash Image model, known as Nano Banana.
3. Upload a clear photo of the subject you want to model (a full-body shot works best).
4. Use a detailed descriptive prompt to generate the figurine. For example, you can use a prompt requesting a 1/7 scale realistic figurine positioned in a real environment with specific details.
5. Click "Generate" to create the 3D figurine image.
6. You can download the generated image.
7. For a full 3D model (beyond the image), use the Gemini-generated image and upload it to an external tool like studio.triple3d.ai to convert the 2D render into a 3D model with textures and rigging.

This workflow uses Gemini to create the 3D figurine's image and external tools to fully generate and export the 3D model itself. The method is fast, typically free, and leverages AI to turn photos into stylized 3D figurines with professional-grade looks without manual modeling.

If a sample prompt is helpful, here is one commonly used: "A highly detailed 1/7 scale figurine of the characters from the provided picture, crafted in a realistic style, is displayed in a real-world environment. The figurine is posed dynamically on a round, transparent acrylic base with no text, placed on a sleek, modern computer desk made of wood with a glossy finish..." This prompt guides the AI to produce a professional-quality 3D figurine image in Gemini.

Let me know if detailed guidance on the prompt or 3D model exporting steps is needed.