Book a Demo

Experience Delhivery Maps

Test core APIs, run live requests, and explore logistics intelligence for free.

Geocoding API

Convert complex Indian addresses into precise geographic coordinates

Reverse Geocoding API

Translate coordinates into verified, local human-readable addresses.

Address Standardisation API

Parse and structure messy Indian address text into organized components.

Routing API

Generate optimal paths adjusted for Indian traffic and vehicle constraints.

Distance Matrix API

Calculate travel times and distances between multiple origins and destinations

Autosuggest API

Get location-aware address/POI suggestions to minimize user typing.

Map Tiles API

Raster map tiles in WebP format for custom map interfaces.

MCP Server

Give your AI agents native access to Delhivery’s mapping services.

Geocoding

Geocoding API converts free-form Indian addresses into precise geographic coordinates (latitude/longitude). Powered by Naksha LLM, it intelligently interprets informal, incomplete, or misspelled address text to deliver accurate map coordinates for any location across India.

Capabilities
  • Naksha LLM Powered: Understands raw address context and automatically completes missing parts.
  • Forward Geocoding: Converts semi-structured text into precise latitude and longitude coordinates.
  • India-Optimised: Trained on billions of delivery addresses to seamlessly master complex, informal local formats.
  • Structured Output: Returns structured coordinates with an error radius for positional confidence.
Geocode Playground
API Ref
Parameters
Enter any Indian address
Geocoded location will appear here

Response

Click on "Run" to view the response

Reverse Geocoding

Reverse Geocoding API converts geographic coordinates into human-readable addresses. By providing a specific latitude and longitude, the API identifies exactly where that point sits on the map in plain language.

Capabilities
  • LLM-Powered Translation: Uses Naksha LLM to transform raw coordinates into clean, recognizable locations.
  • Multi-Level Granularity: Returns structured address components from street-level to state-level.
  • Ranked Accuracy: Provides the most specific and exact address as the primary result.
  • Intelligent Estimation: Identifies the closest addressable location within a defined tolerance.
Reverse Geocoding Playground
API Ref
Parameters
Click map or enter coords

Response

Click on "Run" to view the response

Address Standardisation

The Address Standardisation API validates, corrects, and standardises messy Indian addresses into clean, structured components with a complete entity breakdown. Driven by Naksha LLM, it fixes typos, fills missing fields like pin code or locality, and outputs a normalized address ready for downstream logistics use.

Capabilities
  • LLM-Powered Validation: Uses Naksha LLM to check address components for correctness and completeness.
  • Intelligent Correction: Automatically fixes typos, informal abbreviations, and formatting errors.
  • Smart Standardization: Outputs a structured, clean address with fully normalized fields and a complete breakdown of individual location entities.
  • Deep Enrichment: Contextually fills in missing pin codes, states, or required locality information.
Address Standardisation Playground
API Ref
Parameters
Enter the address as-is, including typos
Standardised location will appear here

Response

Click on "Run" to view the response

Routing

Routing API calculates the most efficient path between two or more points based on specific criteria like distance, time, and vehicle type. It analyzes road networks, traffic conditions, and turn restrictions to provide a sequence of directions that guides a driver from an origin to a destination.

Capabilities
  • Time-Aware Routing: Generates optimised paths for motorcycles, cars, and trucks based on the time of day.
  • Custom Route Modifiers: Allows full control to avoid tolls and highways or calculate the absolute shortest path.
  • Multi-Stop Planning: Seamlessly plans complex journeys with support for up to 10 sequential coordinate stops.
  • Turn-by-Turn Guidance: Generates precise, easy-to-follow navigation instructions for drivers along the route.
Routing Playground
API Ref
Parameters
One latitude & longitude pair per line
Click map to add points or Run

Response

Click on "Run" to view the response

Distance Matrix

The Distance Matrix API returns distances and travel times between many points. It efficiently computes a full matrix of all origin-destination pairs, returning durations in seconds and distances in meters on the fastest route for each element.

Capabilities
  • Many-to-Many Routing: Computes a full N X M matrix of travel times and distances for all origin-destination pairs simultaneously.
  • Vehicle-Specific Matrix: Calculates matrix elements dynamically based on the selected vehicle profile (Car, Motorcycle, Truck) to account for varying speeds and restrictions.
  • Asymmetric Mapping: Accounts for one-way streets, turn restrictions, and U-turns to ensure route metrics differ accurately between A -> B and B -> A.
  • Optimisation Engine Ready: Designed with ultra-low latency to instantly feed downstream vehicle routing (VRP), dispatching, or clustering algorithms.
Matrix Playground
API Ref
Parameters
Run to plot sources (blue) and targets (red)

Response

Click on "Run" to view the response

Autosuggest

The Autosuggest API delivers real-time place predictions - addresses, businesses, POIs, neighborhoods, postcodes, and more - as users type, using fuzzy and partial matching with optional geo-bias. By applying location bias (latitude/longitude), results are prioritized near the specified area first, ensuring locally relevant suggestions. Ideal for typeahead suggestions and landmark discovery.

Capabilities
  • Fuzzy & Partial Matching: Handles user typos and incomplete address queries gracefully in real time.
  • Geo-Biased Results: Prioritizes suggestions near a specified location without rigidly restricting them.
  • Diverse Location Types: Returns predictions across a wide range of categories, including addresses, landmarks, POIs, and neighborhoods.
  • Low Latency Performance: Designed for instantaneous typeahead search.
Autosuggest Playground
API Ref
Parameters
Run to see location markers

Response

Click on "Run" to view the response

Map Tiles

Fetches raster map tiles in WebP format for rendering custom map interfaces. Supports multiple styles and standard slippy-map tile addressing (z/x/y). Build better maps, faster - custom styling and seamless cross-platform deployment help transform raw coordinates into interactive mapping experiences.

Capabilities
  • Cross-Platform Consistency: Delivers the exact same tile set seamlessly across web, iOS, and Android platforms.
  • High Performance: Uses WebP format to ensure rapid map loading speeds and low bandwidth consumption.
  • Rich Visual Detail: Enriches maps with highly detailed points of interest (POIs) and localities.
  • Standard Compatibility: Supports standard slippy-map addressing for easy integration with Leaflet, Mapbox GL, and native SDKs.
Map Tiles Playground
API Ref
Parameters
4 to 20
Preview tile will appear here

Response

Click on "Run" to view the response
Model Context Protocol

One config. Every map tool. Inside your AI.

Connect the Delhivery Maps API suite to your AI workflow using the Model Context Protocol. A single token and URL allow Claude, Cursor, and other MCP clients to auto-discover and execute location tools natively.

6
Tools exposed
1
Get tokenOne-click copy
2
Add configPaste snippet in your client
3
ConnectTools auto-discover
4
Ask awayAgent picks the right tool

Step 1: Get your access token

Your API token is automatically generated from your active browser session. If it has expired, simply sign in again and refresh this page.

Fetch your token

Step 2: Add the MCP server configuration

Copy and paste this snippet into your AI client’s MCP configuration file. Your token from Step 1 is automatically injected into the code below.

MCP URL https://gateway-maps-pub-int.delhivery.com/mcp
~/Library/Application Support/Claude/claude_desktop_config.json
Where this lives
  • Claude Desktop (macOS): ~/Library/Application Support/Claude/claude_desktop_config.json
  • Claude Desktop (Windows): %APPDATA%\Claude\claude_desktop_config.json
  • Claude Code: Run claude mcp add from your terminal - no file edit needed
  • Cursor: Settings > MCP > ~/.cursor/mcp.json
  • Kiro: Command Palette > MCP Configuration > ~/.kiro/settings/mcp.json
What happens after restart
  • Your AI client connects to the MCP server on startup.
  • Tools are auto-discovered - no extra wiring.

Step 3: Available tools

geocode_address

Converts addresses into latitude and longitude coordinates.

"What are the coordinates for Indiranagar, Bangalore?"
standardize_address

Parses unstructured address text into structured fields.

"Standardise this: flat 4b, sec 45, gurgaon"
route

Calculates driving paths between waypoints based on traffic and vehicle types.

"Plan a route from Connaught Place to Gurugram"
compute_distance_matrix

Calculates travel times and distances between multiple origins and destinations.

"Which warehouse is closest to these 3 stops?"
reverse_geocode

Converts latitude and longitude coordinates into addresses.

"What address is at 28.61, 77.23?"
auto_suggest

Finds places and addresses nearby as you type.

"Suggest places matching '1450 Palam Vihar' near 28.51, 77.04"

Client Integration Guides

Already updated your configuration file? Just restart your client. Otherwise, select your AI tool below to get started.

Claude Desktop

macOS · Windows
  1. Open Claude Desktop, click Settings → Developer.
  2. Click Edit Config. The config JSON opens in your editor.
  3. Paste the snippet from Step 2 inside mcpServers.
  4. Save the file and restart Claude Desktop.
  5. Look for the 🔌 icon in the message bar - tools are live.

Claude Code

Anthropic CLI · terminal
  1. Open your terminal where you run claude.
  2. Copy the command from Step 2 (the Claude Code tab).
  3. Paste & run it - it registers a project-scoped server named maps.
  4. Run claude mcp list to confirm maps shows up as connected.
  5. Start a new session - Claude Code auto-discovers all tools.

Cursor

Inline AI editor
  1. Open Cursor Settings → Tools & MCP.
  2. Click New MCP Server.
  3. Paste the JSON snippet (transport: stdio).
  4. Hit save. Cursor will spawn the server automatically.
  5. The tools show up in chat with the @-mention picker.

Kiro

AI-powered IDE
  1. Open Command Palette (⌘⇧P / Ctrl+Shift+P).
  2. Run Kiro: Open MCP Configuration (or open ~/.kiro/settings/mcp.json).
  3. Paste the snippet from Step 2 inside mcpServers.
  4. Save the file - Kiro hot-reloads MCP servers automatically.
  5. Open the MCP Servers panel to verify tools are listed.