Skip to content
Hex Intelligence

India, decomposed into ~530M hexagons. Each one labelled.

Uber H3 grid + GaliGali signal stack: density, mobility, dwell, civic activity, vibe. The atomic unit of FAANG-of-Bharat data.

Cells with ≥1 label
12.4M
Distinct labels
32
Avg labels / live cell
7.1
Refresh cadence
24h

Endpoints

MethodPathCostTier
GET/v1/hex/labels?h3=
All labels & confidences for one cell.
8free
GET/v1/hex/freshness?h3=
When was this cell last touched, and by whom?
4free
GET/v1/intel/hex/:h3
Showcase: labels + admin + transit + POIs + env in one call.
35pro

Try it from your terminal

No SDK install. Plain HTTP. Works the same from a Lambda, a Jupyter notebook, or an LLM tool call.

curl -H "X-API-Key: $GR_KEY" \
  "https://api.gridrock.ai/v1/hex/labels?h3=8861aacd1bfffff" | jq
For agents

Drop into Claude / OpenAI / LangGraph as hex_intel

Bind once. Call on demand. Get back a verbose, deterministic shape that any LLM can reason over without prompt engineering.

# OpenAI function-calling
{
  "name": "hex_intel",
  "arguments": { "h3": "8861aacd1bfffff" }
}
# → returns the kitchen-sink shape an LLM can summarise into "this is a high-density mixed-use cell in Bandra West with strong evening dwell."
Why Bharat

Western hex providers ship US/EU labels. Ours are trained on Indian street-level signals: chai-stall density, autorickshaw dwell, festival flux, monsoon-flood polygons.

Related

Start building with Hex — free

No card. No sales call. 60-second self-serve, agents included.

Get a free key