

A new artificial intelligence-powered art trend is rapidly gaining traction on social media, with creators producing adorable, chibi-style 3D dioramas that resemble miniature collectibles. These images showcase tiny, cartoon-like versions of users in cozy, cube-shaped rooms, rendered in an isometric 3D style with generative AI tools.
The trend has flooded platforms like X, where users are sharing personalised cube scenes that look like desk lamps, toy models, or display figurines. Warm lighting, realistic textures, and small imperfections give the visuals a tactile, almost physical feel.
The trend received a significant boost after AI creator Nano Banana Pro shared a ready-to-use prompt that allows users to generate these scenes in just minutes. The prompt guides users to describe a person, their activity, and the setting, which the AI then converts into a polished 3D cube diorama.
Key elements include an enclosed cube layout, a chibi figurine with oversized head and small body proportions, matte PVC-style materials, and soft internal lighting. According to Nano Banana Pro, the aim is to create “cute little diorama lamps” that feel like real objects rather than flat digital images.
Nano Banana Pro also credited X user @Arminn_Ai for inspiring the prompt structure. Earlier in December, @Arminn_Ai shared examples of miniature isometric rooms and encouraged users to generate personalised versions by uploading photos.
Their expanded prompt dives deeper into scene-building, asking users to define room décor, furniture, clutter, lighting mood, and even the character’s facial expression and pose. A strict isometric viewpoint and cube-contained design ensure the final image maintains a clean, collectible-style appearance.
The popularity of these cube scenes lies in their blend of nostalgia, creativity, and self-expression. Visually, they resemble dollhouses or toy figurines, triggering familiar memories while showcasing the growing sophistication of text-to-image AI tools.
What makes the trend especially appealing is accessibility. Users don’t need professional 3D software or design skills, just a well-written prompt. In minutes, they can create images that look like professionally crafted miniatures.
Also Read: How to Use Gemini Nano Banana Pro to Build a 3D Caricature
As generative AI continues to evolve, trends like chibi-style 3D dioramas highlight how digital creativity is becoming more playful and personal. By combining simple descriptions with powerful AI tools, users are turning everyday ideas into charming miniature worlds that feel both nostalgic and cutting-edge.