Architecture and interior design firms manage enormous volumes of creative and technical documentation: design briefs, material specifications, building code research, client presentations, space planning narratives, and construction documentation. Chinese LLMs like DeepSeek V4, GLM-4, and Qwen3 can accelerate these workflows by generating concepts, checking compliance, drafting specifications, and creating compelling client communications. TokenEase's unified API gives design firms access to these powerful models at a fraction of typical AI costs.
Transform client requirements and site constraints into creative design concepts with programmatic descriptions.
import requests
def generate_design_concept(client_brief, site_constraints, style_preferences):
response = requests.post(
"https://tokenease.io/v1/chat/completions",
headers={"Authorization": "Bearer YOUR_TOKENEASE_API_KEY"},
json={
"model": "deepseek-v4",
"messages": [
{"role": "system", "content": "You are a creative architect. Generate design concepts that respond to briefs and constraints with innovative, practical solutions. Include spatial organization, material palette, and sustainability considerations."},
{"role": "user", "content": f"Brief: {client_brief}\nSite: {site_constraints}\nStyle: {style_preferences}\n\nGenerate 3 distinct design concepts."}
]
}
)
return response.json()["choices"][0]["message"]["content"]
brief = "Mixed-use development: 15,000 sqm, retail ground floor, 8 floors residential above, underground parking"
site = "Corner lot in historic district, height limit 35m, solar access requirements to north neighbor"
style = "Contemporary with contextual references to surrounding Victorian architecture"
concepts = generate_design_concept(brief, site, style)
Automate the verification of design proposals against complex, jurisdiction-specific building codes and regulations.
def check_building_compliance(design_parameters, building_code, jurisdiction):
response = requests.post(
"https://tokenease.io/v1/chat/completions",
headers={"Authorization": "Bearer YOUR_TOKENEASE_API_KEY"},
json={
"model": "qwen3-235b",
"messages": [
{"role": "system", "content": f"You are a building code consultant for {jurisdiction}. Check design compliance and identify any violations or required modifications."},
{"role": "user", "content": f"Building code: {building_code}\nDesign parameters:\n{design_parameters}\n\nCheck compliance and list issues."}
]
}
)
return response.json()["choices"][0]["message"]["content"]
design = """
Building type: Residential apartment, 12 stories
Floor area: 850 sqm per floor
Unit mix: 2-bedroom (65 sqm), 3-bedroom (95 sqm)
Stair width: 1.2m
Corridor width: 1.5m
Ceiling height: 2.7m
Fire rating: 2-hour between units
"""
code = "IBC 2021 Chapter 10 (Means of Egress), Chapter 7 (Fire Resistance), local amendments for residential density"
compliance = check_building_compliance(design, code, "Chicago, IL")
Generate detailed material specifications and suggest sustainable alternatives when preferred materials are unavailable or over budget.
def generate_material_specification(design_element, performance_requirements, budget_constraint):
response = requests.post(
"https://tokenease.io/v1/chat/completions",
headers={"Authorization": "Bearer YOUR_TOKENEASE_API_KEY"},
json={
"model": "glm-4-plus",
"messages": [
{"role": "system", "content": "Generate CSI-format material specifications. When materials are unavailable or over budget, suggest equivalent alternatives with comparable performance."},
{"role": "user", "content": f"Element: {design_element}\nRequirements: {performance_requirements}\nBudget: {budget_constraint}\n\nGenerate spec and alternatives."}
]
}
)
return response.json()["choices"][0]["message"]["content"]
element = "Exterior cladding for high-rise residential, floors 3-12"
requirements = "Fire resistance rating 1-hour, wind load 150 km/h, thermal performance R-15, aesthetic: warm metallic finish"
budget = "$180/sqm installed, preferred material unavailable (zinc panels)"
spec = generate_material_specification(element, requirements, budget)
Craft compelling design narratives that communicate architectural vision to clients who may not have technical backgrounds.
def generate_client_presentation(design_concept, audience, key_selling_points):
response = requests.post(
"https://tokenease.io/v1/chat/completions",
headers={"Authorization": "Bearer YOUR_TOKENEASE_API_KEY"},
json={
"model": "kimi-k2",
"messages": [
{"role": "system", "content": "Write client-facing design presentations. Use accessible language, emphasize benefits over technical details, and create emotional connections to the design."},
{"role": "user", "content": f"Concept: {design_concept}\nAudience: {audience}\nKey points: {key_selling_points}\n\nWrite presentation narrative."}
]
}
)
return response.json()["choices"][0]["message"]["content"]
concept = "Biophilic office design with central atrium, living walls, natural ventilation strategy, and wellness-focused amenities"
audience = "Tech startup founders, ages 28-35, first office buildout, budget-conscious but values employee wellbeing"
points = "30% productivity increase from natural light, 40% reduction in HVAC costs, talent retention"
presentation = generate_client_presentation(concept, audience, points)
Generate optimized space plans based on programmatic requirements, adjacency matrices, and circulation flows.
def optimize_space_plan(program_requirements, adjacency_preferences, constraints):
response = requests.post(
"https://tokenease.io/v1/chat/completions",
headers={"Authorization": "Bearer YOUR_TOKENEASE_API_KEY"},
json={
"model": "deepseek-v4",
"messages": [
{"role": "system", "content": "Generate optimized space plans with room layouts, circulation paths, and area allocations. Consider adjacency requirements and building constraints."},
{"role": "user", "content": f"Program: {program_requirements}\nAdjacencies: {adjacency_preferences}\nConstraints: {constraints}\n\nSuggest space plan."}
]
}
)
return response.json()["choices"][0]["message"]["content"]
program = """
Restaurant: 300 sqm total
- Dining area: 120 seats
- Kitchen: 60 sqm
- Bar: 20 seats
- Restrooms: 2 sets
- Storage: 15 sqm
- Reception/waiting: 20 sqm
"""
adjacency = "Kitchen adjacent to dining. Bar between entrance and dining. Restrooms accessible without crossing kitchen."
constraints = "Column grid 8m x 8m. Windows on north and east only. Existing structural walls on south side."
plan = optimize_space_plan(program, adjacency, constraints)
Automatically generate revision clouds, change narratives, and coordination notes from design change descriptions.
def generate_revision_narrative(change_description, affected_drawings, revision_number):
response = requests.post(
"https://tokenease.io/v1/chat/completions",
headers={"Authorization": "Bearer YOUR_TOKENEASE_API_KEY"},
json={
"model": "qwen3-235b",
"messages": [
{"role": "system", "content": "Generate professional architectural revision narratives following AIA standards. Include scope, reason, and affected disciplines."},
{"role": "user", "content": f"Revision: {revision_number}\nChange: {change_description}\nDrawings affected: {affected_drawings}\n\nGenerate revision narrative."}
]
}
)
return response.json()["choices"][0]["message"]["content"]
change = "Relocate main stair from west to east side of building to improve lobby visibility and natural light penetration. Adjust structural framing accordingly."
drawings = "A-101 Floor Plan Level 1, A-102 Floor Plan Level 2, S-201 Framing Plan Level 1, S-202 Framing Plan Level 2, M-301 HVAC Plan Level 1"
narrative = generate_revision_narrative(change, drawings, "Rev 3")
Get $1 free credits (1M tokens) to generate concepts and automate specifications.
Start designing with TokenEase →