How to Craft Free Action-Figure 3D Models with Google Gemini?

In an era where artificial intelligence is transforming everyday creativity, a quirky new trend is sweeping social media: turning ordinary photos into stunning, collectible-style 3D action figures. Dubbed the “Nano Banana” craze, this viral phenomenon leverages Google’s Gemini AI to create hyper-realistic digital figurines that look like they jumped straight out of a toy store shelf. Whether you’re envisioning yourself as a superhero, your pet as a designer plush, or a favorite character in dynamic pose, the best part? It’s completely free and requires no design expertise. As of September 2025, over 200 million images have been generated using this tool, fueling endless posts on Instagram, X, and TikTok.

This isn’t just a fleeting fad; it’s a gateway to democratizing 3D modeling. Traditional methods demand expensive software like Blender or ZBrush, plus hours of learning curves. But with Gemini’s Nano Banana – officially the Gemini 2.5 Flash Image model – anyone with a Google account and a smartphone can produce professional-grade results in seconds. Launched quietly earlier this year and promoted by Google itself on X, the tool excels at preserving facial likeness, adding intricate details like transparent bases and packaging, and rendering scenes with dramatic lighting. It’s sparked a wave of user-generated content, from fantasy avatars to sports star collectibles, proving AI’s power to make high-end creativity accessible.

What makes Nano Banana stand out?

Its speed and consistency. Unlike earlier AI models that struggled with detail retention, this one handles complex prompts seamlessly, blending uploaded photos with textual instructions for photorealistic outputs. Google has even shared official prompts to kickstart the process, encouraging tweaks for personalization. As Bilawal Sidhu, a former Google PM and AI enthusiast, noted in a viral X thread, Gemini’s multimodal capabilities rival advanced workflows in tools like ComfyUI or Photoshop, all via simple chat-like interactions. But don’t just take my word for it – let’s dive into how you can join the fun.

The Rise of the Nano Banana: From Secret Model to Social Media Sensation

The Nano Banana trend exploded in early September 2025 when AI hobbyists on platforms like LMSYS Arena discovered a mysterious high-performing model that turned photos into lifelike 3D figures. Google soon revealed it as their experimental Gemini 2.5 Flash Image, nicknamed “Nano Banana” for its banana-like efficiency in processing (quick, peel-and-eat simple). The company amplified the buzz with an X post featuring example generations, complete with a ready-to-use prompt that has since been copied millions of times.

Users are raving about its versatility. On X, creators like @Sammyaz01 shared step-by-step breakdowns, emphasizing how it eliminates the need for pro skills: “You don’t need expensive software. All you need is Gemini (and it’s free).”

Meanwhile, @IbanUmar01 posted side-by-side comparisons of before-and-after photos, turning casual selfies into Bandai-style action figures.

The trend’s appeal lies in its therapeutic and shareable nature – imagine gifting a digital “mini-me” to friends or using it for custom merchandise ideas. Even brands are experimenting, creating fantasy avatars to boost engagement on social handles.
Critics might worry about AI’s role in art, but proponents argue it lowers barriers, allowing hobbyists to prototype ideas before investing in physical 3D printing. As one Times of India article put it, “From seeing yourself as a collectible to transforming your pet into a designer toy, Nano Banana’s trend is sweeping creative communities globally.”
With no watermarks or paywalls, it’s a pure expression of innovation.

Why Action-Figure Style? The Allure of Collectible 3D Models

Action figures have long symbolized heroism and imagination, from classic Marvel heroes to modern anime collectibles. Nano Banana taps into this by generating images that mimic high-end toys: semi-realistic sculpts, glossy packaging, and dynamic poses. The outputs aren’t true 3D files (like .obj for printing) but photorealistic 2D renders that simulate 3D depth, perfect for social sharing or as references for actual modeling.

Key features include:
Realistic Detailing: Dramatic shadows, balanced lighting, and material textures like acrylic bases or PVC plastic.
Customization: Mix styles – anime, cartoon, superhero, or sci-fi – for unique twists.
Contextual Scenes: Place your figure on a desk with a mock ZBrush screen and toy box for that authentic “unboxing” vibe.

This style resonates because it’s nostalgic yet futuristic. As Creative Bloq observed, “With the right prompt, the AI model can generate an image that combines the likeness of the reference photo with the look of a detailed toy figure, including any props or accessories.”

Step-by-step Guide: Bring Your Action Figure to Life for Free

Ready to create? Follow this simple, no-cost process using Google Gemini. You’ll need a Google account and a device with internet. The whole thing takes under five minutes.

Step 1: Access Google Gemini AI Studio

Head to gemini.google.com or download the Gemini app from the Google Play Store or Apple App Store. Sign in with your Google account – it’s free, with generous usage quotas for non-commercial use. Once inside, look for the “Try Nano Banana” option or select the Gemini 2.5 Flash Image model from the dropdown. This ensures you’re using the optimized tool for figurine generation.
Accept any terms if prompted, and you’re set.

Step 2: Upload Your Base Photo

Click the “+” or upload icon near the prompt box. Choose a clear, high-resolution photo – a portrait or full-body shot works best for likeness preservation. Selfies, pet pics, or character illustrations all yield great results. Pro tip: Use natural lighting in your photo for better AI interpretation. As @tryagentsea demonstrated on X, this step is straightforward: “Upload a pic of yourself or any image you’d like to use.”

Step 3: Craft Your Prompt for Action-Figure Magic

In the text box, paste a detailed prompt to guide the AI. Start with Google’s official one for reliable results, then customize for action-figure flair.

Here’s the base prompt shared by Google on X:
” a 1/7 scale commercialized figurine of the characters in the picture, in a realistic style, in a real environment. The figurine is placed on a computer desk. The figurine has a round transparent acrylic base, with no text on the base. The content on the computer screen is a 3D modeling process of this figurine. Next to the computer screen is a toy packaging box, designed in a style reminiscent of high-quality collectible figures, printed with original artwork. The packaging features two-dimensional flat illustrations.

For an action-figure twist, tweak it like this: “Create a highly detailed 1/6 scale action-figure style figurine of the character(s) from the uploaded photo in a semi-realistic superhero style. Dynamic pose with dramatic lighting, on a round transparent acrylic base. Place on a computer desk with ZBrush modeling on the screen and a Bandai-style toy box beside it, featuring vibrant 2D artwork.”

Experiment with variations:
Superhero Vibe: “Transform into a superhero action figure, dynamic stance, with a cape and comic book packaging beside the figurine.
Anime Style: “Turn the photo into an anime figurine on a clear acrylic base, vibrant pose, with a manga-style background and neon lighting.”
Pet Collectible: “Make a sports star pet collectible, in a jersey uniform, on a mini stadium, celebration pose, with a trophy accessory.”

The more specific, the better – mention scale (1/6 or 1/7), style (realistic, cartoon), and accessories for polished outputs.

Step 4: Generate and Refine

Hit “Generate” or “Run.” In seconds, Nano Banana processes your input, producing one or more images. Review the results – if it’s not perfect, refine the prompt (e.g., “Make the pose more heroic” or “Add sci-fi elements”) and regenerate. Gemini supports conversational editing, so you can chain prompts for iterations, like adding “Enhance the shadows for more depth.”
Download your favorites in high-res.

Step 5: Share and Iterate

Post your creation on social media with hashtags like #NanoBanana, #AIFigurine, or #ActionFigureTrend. For physical versions, use the image as a reference for 3D printing services. As @princeofworld45 shared on X, it’s as easy as 1-2-3-4 steps to get started.

Pro Tips for Jaw-Dropping Results

Photo Quality Matters: Opt for front-facing, well-lit images to maintain accuracy.
Prompt Engineering: Use descriptive language – “photorealistic,” “dramatic lighting,” “Bandai-style packaging” – to elevate outputs. Combine styles like “hologram-model figure with transparent lines and sci-fi aesthetic.”
Avoid Overloading: Start simple; complex scenes can sometimes glitch.
Ethical Note: Respect likeness rights when using others’ photos.
Limitations: Outputs are images, not editable 3D models. For true 3D files, export to tools like Blender post-generation.
Troubleshooting: If access is limited, check your region’s availability or usage quotas.

Beyond the Trend: The Future of AI-Driven Creativity

The Nano Banana trend highlights AI’s potential to revolutionize hobbies and industries. From merchandising to education (visualizing historical figures as figurines), it’s opening doors. As PetaPixel noted, “Google’s Nano Banana model… has got its first fad,” but it’s just the beginning. Imagine integrating this with Veo for animated figures or using it in game dev for quick prototypes.

In conclusion, creating action-figure style 3D models with Google Gemini is a thrilling, barrier-free entry into digital art. It’s not about replacing artists but empowering everyone to play. So, grab your photo, fire up Gemini, and unleash your inner creator – your viral moment awaits.

Naorem Mohen

Naorem Mohen is the Editor of Signpost News. Explore his views and opinion on X: @laimacha.