AI Elevation with Specific Materials: Prompting for Stone, Glass, Wood & More

Why Your AI Elevation Material Prompt Matters More Than You Think

If you have tried generating a facade design using AI, you have likely faced the same frustration: the output looks generic. The walls appear flat, the textures are vague, and the materials resemble nothing you would actually find at a stone dealer in Rajasthan or a tile showroom in Morbi. The reason is almost always the same — the ai elevation material prompt was too vague. Writing “beautiful modern house” tells the AI nothing about whether you want Kota stone, Saint-Gobain glass, or teak wood battens on your facade. In this guide, we break down exactly how to specify Indian building materials in your prompts so that Elevations by Ongrid Design produces outputs your contractor can actually work from. If you are new to prompt writing for elevations, start with our 10 prompt formulas that generate stunning house elevations before diving into material-specific techniques.

Understanding How AI Reads Your AI Elevation Material Prompt

The Elevations tool interprets material prompts along four axes:

• Material type — stone, glass, wood, metal, composite
• Finish or texture — polished, honed, bush-hammered, matte, reflective
• Pattern or layout — stacked bond, running bond, random ashlar, vertical battens
• Colour tone — warm beige, charcoal grey, honey brown, off-white

A prompt that covers all four axes will always outperform one that only mentions the material type. For example:

Weak prompt: “stone facade house elevation”

Strong prompt: “Front elevation of a 40x60 duplex with bush-hammered Tandur grey limestone in random ashlar pattern, with a recessed entrance framed in warm teak wood battens”

The strong prompt gives the AI enough information to generate texture, shadow depth, joint patterns, and colour contrast — all of which make the output look like a photograph rather than a cartoon.

The Role of Indian Material Names

A common concern is whether the AI understands Indian-specific material names. Terms like “Kota stone,” “Kadappa black,” “Jodhpur sandstone,” “Sadarahalli granite,” and “Burma teak” are well-represented in architectural training data and are handled reliably. However, hyper-local trade names — such as a specific quarry’s marketing name for a granite variant — may not register. When recognition is medium, the output does not break; it simply becomes more generic, defaulting to a visually plausible but less distinctive stone. In those cases, describe the material by its visual properties instead.

Indian Material Name	AI Recognition	Recommended Prompt Strategy
Kota stone (blue/green)	High	Use “Kota blue limestone, honed finish”
Jodhpur sandstone	High	Use “Jodhpur golden sandstone, split-face”
Kadappa black limestone	High	Use “Kadappa black limestone, polished”
Sadarahalli granite	Medium	Add “light grey granite with subtle grain”
Bansi Paharpur pink sandstone	Medium	Add “pink sandstone, hand-dressed blocks”
Makrana white marble	High	Use “Makrana white marble, polished slabs”
Tandur grey limestone	Medium-High	Add “grey limestone, bush-hammered texture”

You can source most of these materials through dealer clusters like Kishangarh (marble), Kota (limestone), and Morbi (tiles), or through platforms like IndiaMart for pan-India delivery.

AI Stone Facade Prompt: Getting Natural Stone Right

Natural stone is the most popular facade material for independent houses across India — from the sandstone-clad bungalows of Jaipur to the granite-fronted villas of Bengaluru. It is also the material that benefits most from precise prompting, because the difference between “stone wall” and a properly specified ai stone facade prompt is enormous.

Prompt Structure for Stone Facades

Use this template:

“[Stone type] + [finish] + [laying pattern] + [joint treatment] + [application area]”

Example prompts:

1. “Ground floor facade clad in split-face Jodhpur sandstone in a coursed random pattern with 10mm recessed joints, upper floor in smooth off-white exterior plaster”

2. “Compound wall and entrance porch in Kota blue limestone, honed finish, stacked horizontal courses with flush pointing”

3. “Feature wall on the front elevation in Tandur grey limestone, bush-hammered texture, large-format 600x300mm slabs with 5mm shadow joints”

Stone Finish Terminology That Works

The finish you specify dramatically changes how light interacts with the surface in the generated elevation:

Finish Term	Visual Effect	Best For
Polished	Reflective, glossy, shows veining	Marble feature walls, granite dados
Honed	Smooth but matte, subtle texture	Limestone facades, Kota stone flooring
Bush-hammered	Rough, dimpled, strong shadow play	Granite and limestone accent walls
Split-face	Naturally broken surface, rustic	Sandstone boundary walls, plinths
Flamed / Thermal	Rough but flat, anti-slip	Granite cladding, outdoor steps
Hand-dressed / Chisel-cut	Artisan rough, traditional	Heritage-style Rajasthani facades

Cost Context for Stone Facades

These figures help you match your AI-generated design to your actual budget:

Stone Type	Indicative Cost (₹ per sq ft, supply only)	Common Source
Jodhpur sandstone (30mm)	₹45–₹80	Rajasthan
Kota blue limestone (25mm)	₹35–₹55	Kota, Rajasthan
Tandur grey limestone (20mm)	₹40–₹65	Telangana
Kadappa black (20mm)	₹50–₹80	Andhra Pradesh
Sadarahalli granite (20mm)	₹55–₹90	Karnataka
Makrana marble (20mm)	₹120–₹250	Rajasthan
Mint sandstone (30mm)	₹40–₹70	Rajasthan

Installation adds ₹30–₹60 per sq ft depending on city and pattern complexity. In Pune, Hyderabad, and Ahmedabad, skilled stone masons charge ₹600–₹900 per day.

AI Glass Elevation: Specifying Glazing Systems Accurately

Glass is the second most misrepresented material in AI elevations. A prompt that says “glass facade” will produce a generic curtain wall resembling a commercial office — not what you want for a residential project in Gurugram or Chennai. The key to a realistic ai glass elevation is specifying the glazing system, not just the material.

Types of Glass Systems for Indian Residences

System Type	Prompt Description	Typical Use
Aluminium-framed windows	”Powder-coated dark grey aluminium window frames with clear glass”	Most common for houses in the ₹80 lakh–₹2 crore range
Slim-profile steel frames	”Black slim steel window frames with fixed glass panels”	Contemporary villas, ₹2 crore and above
Frameless glass railing	”12mm toughened frameless glass railing with stainless steel standoffs”	Balconies, terrace edges
Glass curtain wall (structural glazing)	“Structural glazing with spider fittings and 12mm toughened glass panels”	Double-height living rooms, staircase walls
Louver windows with glass	”Glass louver windows in anodised aluminium frames”	Bathrooms, utility areas, common in Kerala

Glass Cost Comparison

Glass System	Indicative Cost (₹ per sq ft, supply + installation)	Key Brands
Aluminium-framed windows (5mm clear glass)	₹250–₹400	Fenesta, Tostem, Kommerling
Slim-profile steel frames (6mm clear glass)	₹500–₹800	Custom fabrication, Crittall-style
Frameless glass railing (12mm toughened)	₹350–₹550	AIS, Saint-Gobain, Modiguard
Structural glazing (12mm toughened, spider fittings)	₹700–₹1200	Saint-Gobain, AIS
Glass louver windows (anodised aluminium frame)	₹200–₹350	Betavent, Alteza

Prompt Examples for Glass Elevations

1. “Front elevation of a 3-storey contemporary house with a double-height structural glazing panel over the staircase, dark grey aluminium-framed casement windows on bedrooms, and a frameless glass railing on the first-floor balcony”

2. “South-facing elevation with solar control tinted glass in dark bronze aluminium frames, horizontal sun shading louvres above each window opening”

3. “Minimalist white box elevation with large fixed glass panels in black slim steel frames, floor-to-ceiling glazing on the ground floor living area”

For homeowners in hot climates — Nagpur, Vijayawada, Ahmedabad — specify performance glass. Terms the AI recognises include “tinted glass” (grey, bronze, green), “reflective glass,” “frosted / acid-etched glass,” and “high-performance double-glazed units.” Mentioning Indian brands like Saint-Gobain or AIS can steer the output towards realistic residential glazing rather than European-style triple-glazed units.

AI Wood Cladding Design Prompt: Timber and Composites

Wood on an Indian facade is almost always solid hardwood battens, engineered wood panels, or WPC (Wood Plastic Composite) cladding. The challenge with an ai wood cladding design prompt is that the AI defaults to Scandinavian or North American aesthetics unless you steer it towards Indian species and installation styles.

Prompt Template for Wood Facades

“[Wood type or composite] + [profile: battens / planks / panels] + [orientation: vertical / horizontal] + [spacing] + [colour tone] + [application area]”

Example prompts:

1. “Vertical Burma teak battens (50x25mm) at 25mm spacing over the entrance porch, oiled natural honey-brown finish, mounted on a concealed MS frame”

2. “Horizontal WPC cladding planks in walnut brown on the first-floor front wall, with a dark grey aluminium edge trim”

3. “Thermowood ash cladding in charred black finish (Shou Sugi Ban style) on the side elevation feature wall, vertical orientation”

Wood vs WPC vs HPL: What to Prompt For

Material	Prompt Keyword	Look	Durability	Cost (₹ per sq ft, installed)
Burma teak (solid)	“solid teak wood battens”	Rich grain, warm tone	25+ years with maintenance	₹350–₹600
Thermowood (treated pine/ash)	“thermally modified wood cladding”	Uniform, dark brown	15–20 years	₹250–₹400
WPC (Auralex, Hardy Plast)	“WPC composite wood cladding”	Uniform, less grain	15–20 years, low maintenance	₹120–₹200
HPL (Fundermax, Trespa)	“HPL exterior cladding panels”	Flat, precise, modern	20+ years	₹200–₹350
Engineered bamboo	”bamboo cladding panels”	Linear grain, light	10–15 years	₹150–₹250

If you are building in a coastal city like Mumbai, Kochi, or Visakhapatnam, WPC or HPL is usually the practical choice over solid wood due to salt air and humidity. For a deeper dive into HPL specifically, see our guide on HPL cladding for elevation: pros, cons, cost and Indian brand options. Your prompt should reflect this reality so the generated elevation matches what you will actually build.

How to Get Realistic Texture in AI Elevations

Achieving a realistic texture in an AI-generated facade is a cross-cutting skill that applies to every material — stone, glass, and wood alike. Without texture cues, the AI defaults to flat colour fills that look like a painted diagram rather than a constructed building.

Stone Texture Tips

For natural stone, specify the surface treatment and how light reveals it:

• “Bush-hammered Tandur limestone with deep pock marks catching afternoon sun from the west”
• “Split-face sandstone with visible chisel marks and natural colour variation from cream to amber”
• “Polished Kadappa black with mirror-like reflections and visible fossil inclusions”

Glass Texture Tips

Glass texture is about reflections, transparency, and frame shadow:

• “Clear glass panels reflecting the surrounding tree canopy and sky”
• “Tinted bronze glass with visible interior silhouettes, aluminium mullion shadows”
• “Frosted acid-etched glass glowing with interior backlight at dusk”

Wood Texture Tips

The secret to realistic wood is grain direction, shadow gaps, and lighting interaction:

• “Visible wood grain texture with shadow gaps between battens”
• “Warm afternoon sunlight raking across the wood surface”
• “Natural weathering patina on oiled teak” (for a lived-in look)

Stone vs Tile in AI Elevations: A Common Confusion

One of the most frequent questions from homeowners using Elevations by Ongrid Design: “How do I get the AI to show tiles that look like stone, versus actual stone?” The distinction matters because natural stone and stone-look tiles have very different visual characteristics — and very different costs. For a broader comparison across all exterior cladding options available in India, read our complete guide to elevation cladding materials.

Visual Differences the AI Can Capture

Feature	Natural Stone	Stone-Look Tile
Surface variation	High — each piece unique	Low — pattern repeats every 3–4 tiles
Joint width	8–15mm typical	2–3mm typical (rectified edges)
Edge profile	Irregular, hand-cut feel	Machine-straight, uniform
Thickness	20–40mm	8–12mm
Shadow depth at joints	Deep, visible	Minimal
Colour variation	Significant within a single lot	Controlled, consistent

Prompt Strategies

For natural stone appearance:

“Facade clad in natural Jodhpur sandstone with visible thickness at edges, wide 12mm recessed joints, each stone piece showing unique colour variation from cream to golden”

For stone-look porcelain tile:

“Exterior wall finished in large-format 600x1200mm porcelain tiles with a stone-effect surface, uniform grey tone, minimal 2mm joints, perfectly flat plane”

For exposed stone masonry (load-bearing):

“Traditional Chettinad-style exposed stone masonry wall with lime mortar pointing, random rubble pattern, mixed grey and brown tones”

The key differentiators in your prompt are joint width, surface variation, and edge regularity.

Cost Comparison: Stone vs Tile

Parameter	Natural Stone (Jodhpur sandstone)	Stone-Look Porcelain Tile
Material cost	₹45–₹80 per sq ft	₹60–₹120 per sq ft
Installation cost	₹40–₹60 per sq ft	₹25–₹35 per sq ft
Total installed	₹85–₹140 per sq ft	₹85–₹155 per sq ft
Weight on wall	45–55 kg per sq m	18–22 kg per sq m
Maintenance	Sealing every 2–3 years	Virtually nil
Best for	Ground floor, compound walls, plinths	Upper floors, large flat surfaces

In many Indian cities — Lucknow, Indore, Chandigarh — homeowners combine both: natural stone on the ground floor plinth and entrance, with stone-look tiles on upper storeys to reduce structural load and cost. Your ai elevation material prompt can reflect this mixed approach for a more buildable result.

Advanced Material Prompting Techniques

Once you are comfortable with individual materials, you can combine them for sophisticated multi-material facades. The AI handles material transitions, shadow interactions, and colour harmony across different surfaces effectively when given clear instructions.

The Material Palette Approach

Instead of prompting for one material at a time, describe a complete material palette:

“Contemporary duplex elevation with the following material palette: (1) Ground floor plinth in split-face Kota stone up to sill level, (2) Main walls in off-white textured exterior paint with Birla White cement base, (3) Staircase tower clad in vertical thermowood battens in honey brown, (4) All windows in dark grey powder-coated aluminium frames with clear glass, (5) Balcony railings in MS flat bars painted charcoal black, (6) Parapet capping in matching grey Kadappa stone”

This kind of comprehensive prompt produces elevations that feel designed rather than generated — because every surface has been intentionally specified.

Budget Tier to Material Palette Mapping

Before writing your prompt, match your budget to a realistic material palette:

Budget Tier	Facade Budget (30x40 plot)	Primary Material (60–70%)	Secondary Material (20–30%)	Accent (5–10%)
Economy	₹2–4 lakh	Textured exterior paint (Birla Putty / Asian Ace)	Kota stone plinth or ceramic tiles	MS railing, painted
Mid-range	₹4–8 lakh	Stone-look porcelain tile or sandstone cladding	WPC or HPL cladding on feature wall	Aluminium-framed glazing, SS railing
Premium	₹8–15 lakh+	Full natural stone cladding (Jodhpur / Kadappa)	Solid teak or thermowood battens	Structural glazing, slim steel frames

Specifying Material Transitions

The junction between two different materials is where most AI elevations fail. Fix this by explicitly describing transitions:

• “Stone cladding ending with a clean aluminium edge trim at the junction with plastered wall”
• “Wood battens recessed 50mm behind the main wall plane to create a shadow reveal”
• “Glass railing meeting the stone plinth with a 20mm stainless steel base channel”

Climate-Specific Material Choices

Your city’s climate should inform your material prompt:

Climate Zone	Recommended Facade Materials	Prompt Notes
Hot-dry (Jaipur, Jodhpur, Ahmedabad)	Sandstone, lime plaster, minimal glass	Specify thick walls, deep window reveals, jaali screens
Hot-humid (Mumbai, Chennai, Kochi)	WPC cladding, porcelain tile, treated wood	Mention rain protection details, drainage channels
Composite (Delhi, Lucknow, Nagpur)	Stone + plaster combination, aluminium-glass	Include both sun shading and rain protection
Temperate (Bengaluru, Pune, Mysuru)	Natural stone, exposed brick, large glazing	Specify generous window sizes, ventilation openings
Cold (Shimla, Manali, Gangtok)	Stone masonry, timber, insulated glass	Mention double glazing, thick stone walls, sloped roofs

Common Mistakes in AI Elevation Material Prompts

After analysing thousands of elevation generations on Elevations by Ongrid Design, these are the most frequent material prompting errors:

Mistake 1: Using Generic Colour Names

Wrong: “brown stone wall” Right: “Jodhpur golden sandstone with warm amber and cream tones”

Mistake 2: Forgetting Scale References

Wrong: “stone cladding on the front wall” Right: “300x600mm stone cladding panels on the 12-foot-high front wall”

Mistake 3: Ignoring Joints and Reveals

Joints, shadow gaps, and reveals are what make a material look real. Always specify joint width (2mm for tiles, 8–15mm for stone), joint colour (grey cement, white, dark), and whether joints are flush, recessed, or raised.

Mistake 4: Mixing Incompatible Materials

The AI will render whatever you ask for, even if the combination looks terrible in real life. Avoid high-gloss marble next to rustic split-face stone, multiple competing wood tones, or more than three distinct cladding materials on one facade.

Mistake 5: Not Specifying Light Condition

Materials look completely different under morning sun versus overcast sky versus artificial night lighting. Add a lighting cue: “morning sunlight from the east raking across the textured stone facade” or “warm LED uplighting on the wood-clad entrance porch at dusk.”

Putting It All Together: A Complete Material Prompt Workflow

Step 1: Define your budget range. Use the budget tier table above to identify which materials are realistic for your project. A ₹3–5 lakh facade budget on a 30x40 plot means painted plaster with stone accents — not full stone cladding.

Step 2: Pick your material palette. Choose a primary material (60–70% of the facade), a secondary material (20–30%), and an accent (5–10%).

Step 3: Write the full prompt. Include plot size, number of floors, all materials with finishes and patterns, window systems, railing types, and lighting condition.

Step 4: Iterate and validate. If a specific material looks off, refine just that description in your next prompt. Then share the AI-generated elevation with your mason or contractor — ask whether the stone pattern, tile layout, or wood batten spacing is achievable within your budget. In Bengaluru, Chennai, and Hyderabad, skilled stone cladding contractors are relatively easy to find. In tier-2 cities, you may need to simplify. Ready to try these material prompts yourself? Generate your own elevation and see the difference precise material descriptions make.

The gap between an AI-generated elevation and built reality is almost entirely a function of how well you specified materials in your prompt. Master this skill, and the output becomes genuinely useful — not just for visualisation, but as a working document your architect and contractor can reference throughout construction.

Frequently Asked Questions

How do I specify materials in an AI elevation prompt?

Cover four axes in every prompt: material type (e.g., Jodhpur sandstone), finish (e.g., bush-hammered), pattern (e.g., random ashlar), and colour tone (e.g., warm golden). The more axes you address, the more realistic the output. Use the prompt templates in each material section above as starting points.

Does the AI understand Indian materials like Kota stone or Kadappa black?

Yes. Widely used Indian materials — Kota stone, Kadappa black, Jodhpur sandstone, Makrana marble, Burma teak — have high recognition. For less common materials with medium recognition, supplement the name with a visual description (e.g., “Sadarahalli granite — light grey with subtle grain”).

How do I get realistic texture instead of flat surfaces?

Always specify the surface finish (polished, bush-hammered, split-face) and add a lighting condition (morning sun, overcast, dusk). Mentioning shadow gaps, grain direction, and surface imperfections pushes the AI to generate convincing depth rather than uniform colour fills.

What is the difference between stone and tile in AI-generated elevations?

Prompt for stone by specifying wide joints (8–15mm), natural colour variation, and irregular edges. Prompt for stone-look tile by specifying narrow joints (2mm), uniform tone, and large-format dimensions (600x1200mm). The AI captures these distinctions clearly when you provide the right cues.

Can I combine multiple materials in a single AI elevation prompt?

Absolutely. Use the material palette approach: describe a primary material, a secondary material, and an accent, each assigned to specific zones of the facade. Explicitly describe transitions between materials (edge trims, shadow reveals, base channels) for the most realistic result.