Miniature Country House Diorama Prompt for Gemini, Grok, and GPT Image 2.0

AI visual creator SajidHassain (@SajidHassain6w) has shared a production-grade master prompt and practical workflow on X for generating highly detailed, handcrafted architectural miniature dioramas across visual AI models such as Gemini, Grok, and GPT Image 2.0.

Image source: @SajidHassain6w via X

The technique covers the complete assembly of a charming two-story countryside home on an empty miniature site, featuring cream-colored plaster walls, terracotta roof tiles, wooden shutters, flower boxes, a stone driveway, and a tiny classic red vintage car parked inside an attached garage. Below is the complete prompt structure, step-by-step workflow, and execution tips for both still and video generation.

4-Step Diorama Creation Workflow and Master Prompt

The author outlines a direct four-step procedure designed to leverage reference imagery and dense descriptive prompting:

1. Open your visual AI interface: Access Gemini, Grok, or GPT Image 2.0 inside ChatGPT.
2. Upload a reference photo (optional): Upload a favorite architectural or landscape photo (📸) to guide palette and style.
3. Copy the master prompt: Copy the detailed 9:16 miniature prompt below (⬇️).
4. Generate and customize: Paste the prompt into the generation box and customize individual details for your own variations (✨).

Use the full English master prompt below to generate high-fidelity, handcrafted architectural miniature scenes:

Create a highly realistic cinematic 9:16 miniature diorama video showing the step-by-step construction of a beautiful detailed miniature country house with a working garage.

Begin with an empty miniature construction site and gradually build a charming two-story countryside home. Show realistic miniature craftsmanship as the foundation, walls, doors, windows, wooden shutters, flower boxes, and terracotta roof tiles are carefully constructed and assembled.

The house should have warm cream-colored exterior walls, a traditional terracotta tiled roof, brown wooden shutters, a small chimney, detailed windows, a wooden front door, and an attached garage containing a tiny classic red vintage car.

Show close-up macro shots of hands working on the miniature model, placing tiny bricks, stones, wooden pieces, roof tiles, windows and decorative details. Every construction step should look physically realistic and handcrafted.

Build a beautiful miniature garden around the house with green grass, small trees, colorful flowers, flower pots, stone retaining walls, wooden fencing and a curved stone-paved driveway leading toward the garage.

Use realistic miniature materials and textures: clay roof tiles, plaster walls, natural wood, bricks, stones, soil, grass and tiny plants. Add subtle imperfections and realistic scale to make it look like a genuine handmade architectural model.

Use cinematic macro photography, shallow depth of field, smooth camera movements, slow push-ins, overhead and close-up shots, natural daylight, realistic shadows, soft focus transitions and detailed textures.

As the construction finishes, reveal the complete miniature house and garden in a beautiful cinematic hero shot. Turn on the warm interior lights and garage lights, creating a warm, cozy evening glow against the dusk sky.

The prompt grounds the generation by explicitly specifying physical components—two-story cream walls, clay tiles, wooden window trim, and a tiny classic automobile—providing the model with solid anchor points for handcrafted miniature realism.

Prompt Breakdown and Production Caveats

This prompt succeeds by systematically defining physical materials, miniature scale, and optical characteristics rather than relying on generic aesthetic buzzwords:

• Tangible Material Textures: Calling out specific physical miniature craft materials—clay tiles, plaster, natural timber, mortar, miniature brickwork, soil, and mossy grass—along with intentional subtle imperfections prevents a sterile CGI finish and establishes handcrafted realism.
• Cinematic Macro Photography Grammar: Directives for shallow depth of field, macro push-in camera movement, overhead angles, and soft focus transitions reproduce the optical depth of field characteristics inherent to real-world tilt-shift and macro lens photography.
• Hero Shot Lighting Transition: Transitioning from soft natural daylight to an illuminated evening reveal—activating interior and garage lights against twilight—yields a dramatic, cozy contrast in the final hero view.

When deploying this prompt in production workflows, keep three practical considerations in mind:

1. Text-to-Image Output Behavior: In standard image models like Gemini, Grok, and GPT Image 2.0, the prompt will produce a fully assembled, ultra-detailed macro still photograph of the finished diorama rather than an animated time-lapse sequence.
2. Two-Stage Video (I2V) Production: To achieve animated construction and dynamic camera motion matching the original demonstration clip, feed the generated still image into dedicated Image-to-Video models (such as Kling, Seedance, Runway Gen-3, or Luma Dream Machine) with motion brush and camera path controls.
3. Handling Hand Distortion: Directing the model to depict hands placing miniature bricks can occasionally trigger anatomical or scale artifacts in certain image models. For pristine architectural showcases, remove the hand reference to generate a pure diorama landscape, or clean minor contact artifacts using localized inpainting.

Original source

This practical guide is based on the original workflow and architectural prompt shared on X by creator SajidHassain (@SajidHassain6w). For further implementation details, demonstration videos, and parameter variations across Gemini, Grok, and GPT Image 2.0, refer to the verified links below:

• SajidHassain on X (@SajidHassain6w)
• SajidHassain Profile and Prompt Thread