Infrared City — the climate intelligence layer for buildings and cities

Homepage prototype — not the live site.

A design prototype of the new infrared.city homepage: the story, the order of the sections and the language. Nothing here is wired to the product.

Not finished The 3D film in the hero and the simulation explorer in section 01. Both already run real measured Hong Kong data — 45 floors of daylight, facade solar and sky view, ground heat, wind and sun — but the look, the camera and the interaction are still in work.
Not optimised The scene is about 40 MB and can take ten seconds to appear. Performance is a separate pass. This round is about the storytelling.
One page, many audiences The I am… control switches the persona: the default tab, the questions and the calls to action all change, on the same page. In production the URL does this (?icp=… or /c/…).
Prices are the verified current plans — except Pro, which is still under review. Please do not quote it outside the company.

01 · What it is

Climate you can measure,
before anything is built.

We simulate the climate around and inside buildings — wind, heat, sun and daylight — at metre resolution, in seconds.

Interior daylight · per floor

4.55 %

Daylight factor · floor 30

How much daylight actually reaches a desk on this floor, once the neighbours block the sky. Resolved floor by floor.

0%26%

SOLVE 0.8 S

02 · The differentiator

Buy the layer.
Don't build one.

Everyone else ships an application. We ship the capability — so a product, a practice or a city becomes climate-aware without hiring a simulation team.

Building this yourself means CFD engineers, training data, a GPU cluster, a validation programme — and years. One API call instead:

Climate physics, one API call away. Tiling, polling and merging are handled for you — you get a field back, not a job queue to babysit.

REST API OGC-compliant. Call it from any language or stack.
Python SDK Five lines to a result — tiling, polling, merging handled.
AI-agent skill One-click install for Claude Code, Cursor, Copilot, Windsurf.

# pip install infrared-sdk
from infrared_sdk import Client

client = Client(api_key)
run = client.run_area(
    polygon=aoi,
    analysis="thermal-comfort")

heat = run.result()   # → a field, in seconds

Get an API key Read the docs

Open it as a guest and walk through five public case studies — before you sign up for anything.

Analyze Draw a box, run, read the results in 3D.
Compare Scenarios side by side, per metric.
AI reports Stakeholder-ready, straight from a run.
Integrations & API Plugin downloads and API keys, same account.

Start free See a workflow

900 tokens on signup, 200 every month · no credit card

The engine inside the tools your team already has open — run analyses without leaving your model.

GRASSHOPPER	plugin · Rhino 8 script components	Live
REVIT	add-in	Live
QGIS	plugin	Live
ARCGIS	connector	Live
SKETCHUP	extension	Soon
AGENT SKILLS	Claude Code · Cursor · Copilot	Docs

Get the plugins Tutorials

When the deployment has to fit you rather than the other way round — a city-scale layer, a bespoke workflow, or the engine inside your own environment.

City deployments A whole territory prepared once, served continuously.
Bespoke workflows Your standard, your report format, your acceptance criteria.
Co-development We build the piece that is missing, with your team.

Book a scoping call Who it’s for

04 · Validation

Fast, and honest about it.

0.3 s
Wind
4 s
Ped. comfort
< 5 s
Thermal
~20 s
35-tile district

Model	Measured against	Status
Thermal comfort	2–4 K RMSE on temperate built-up scenes, no per-site calibration	Validated
Wind	High agreement against held-out CFD; benchmarked against AIJ wind-tunnel measurements	Screening
Daylight sky model	Byte-for-byte identical to the Radiance reference	Exact
Per-model studies	Underway; a consolidated overview is in preparation	In progress

Screening-grade. Good enough to compare options and find problems — not a certified measurement. We say so because it is what makes the rest trustworthy.

On one August afternoon in Hong Kong, reads measurably kinder than two blocks over, while the tower's has stored a working day of sun. Touch a noun — the number is measured, not illustrated.

06 · Pricing

Run something today.

One token currency across platform, plugins and API · start without a credit card · paid plans billed annually

	Free	Starter	Advanced · most used	Pro
Per year	€0	€240/yr	€1,200/yr	€6,000/yr
Scale	Students & small studies	Small sites	Quarter scale	District to city
Tokens / month	900 on signup + 200/mo	500	4,000	50,000
≈ analyses / month	~20 every month	50	400	5,000
Price per extra token	€0.10	€0.04	€0.025	€0.01
Access	Platform · plugins · API	Everything · 10 req/s	Everything · 30 req/s	Everything · 60 req/s
Extras	Same engine, same account	All simulation models	Compare + AI reports	Highest throughput
	Start free	Choose	Choose	Choose

Every plan buys top-up packs at its own discount — list rate €0.10 per token · one analysis = 10 tokens · pay only for the area you study

Something bigger?

Custom digital twin · live layer for your city · co-development

Talk to us

Homepage prototype — not the live site.

The climate intelligence layer for buildings and cities.

Climate you can measure,
before anything is built.

Buy the layer.
Don't build one.

Pick the one that sounds like your week.

I'm building a product

I run a city

I produce evidence

I design buildings

Fast, and honest about it.

A whole city, updated through the day.

Run something today.

Homepage prototype — not the live site.

The climate intelligence layer for buildings and cities.

Climate you can measure,before anything is built.

Buy the layer.Don't build one.

Pick the one that sounds like your week.

I'm building a product

I run a city

I produce evidence

I design buildings

Fast, and honest about it.

A whole city, updated through the day.

Run something today.

Climate you can measure,
before anything is built.

Buy the layer.
Don't build one.