A self-hosted navigation, mapping, and location intelligence platform. Render vector maps, compute real-time routes, geocode addresses, and build location-aware applications — all without third-party API dependencies. Powered by OpenStreetMap data with commercial-grade performance.
import kmap
let map = kmap.Map({
center: [-23.5505, -46.6333],
zoom: 13,
style: "koder://streets-v2"
})
// Calculate route with traffic
let route = kmap.directions({
origin: [-23.55, -46.63],
destination: [-23.58, -46.68],
mode: "driving",
traffic: true
})
map.addRoute(route, color: "#16a34a")
Everything you need, built from the ground up.
GPU-accelerated vector map rendering with custom styles, 3D buildings, terrain elevation, and smooth 60fps pan/zoom on all platforms.
Multi-modal routing engine supporting driving, walking, cycling, and public transit with live traffic integration and ETA prediction.
Forward and reverse geocoding with fuzzy search, autocomplete, and address parsing for 200+ countries and territories.
Define virtual boundaries and receive real-time alerts when devices enter, exit, or dwell within geofenced areas.
Download map regions for fully offline navigation. Compact vector tiles use 10x less storage than raster alternatives.
Real-time vehicle and asset tracking with historical playback, route optimization, and delivery zone management.
Embed rich, interactive vector maps with custom styling and data layers.
let map = kmap.Map({ container: "map" })
map.addLayer({
type: "heatmap",
source: "earthquakes",
paint: {
intensity: ["interpolate", ["linear"], ["zoom"], 0, 1, 9, 3],
color: ["interpolate", ["linear"], ["heatmap-density"],
0, "transparent", 0.5, "#16a34a", 1, "#dc2626"]
}
})
Convert addresses to coordinates and search for places with autocomplete.
// Forward geocode
let results = kmap.geocode("Av Paulista 1578, Sao Paulo")
// => { lat: -23.5614, lon: -46.6558, confidence: 0.97 }
// Reverse geocode
let address = kmap.reverse(-23.5614, -46.6558)
// => "Av. Paulista, 1578 - Bela Vista, São Paulo"
// Autocomplete search
let suggestions = kmap.autocomplete("coffee near me",
proximity: [-23.56, -46.65]
)
Create virtual boundaries and monitor device positions in real time.
let fence = kmap.geofence({
name: "warehouse-zone",
geometry: kmap.circle([-23.55, -46.63], radius: 500),
events: ["enter", "exit", "dwell"]
})
fence.on("enter") |device| {
notify.send("Vehicle {device.id} entered zone")
}
See how Koder Map stacks up against the competition.
| Feature | Koder Map | Google Maps | Mapbox | OpenStreetMap | HERE Maps | TomTom |
|---|---|---|---|---|---|---|
| Self-hosted / no API keys | ✓ | — | — | ✓ | — | — |
| Vector tile rendering | ✓ | ✓ | ✓ | Partial | ✓ | ✓ |
| Real-time traffic routing | ✓ | ✓ | ✓ | — | ✓ | ✓ |
| Offline map support | ✓ | Partial | ✓ | Partial | Partial | Partial |
| Built-in geofencing | ✓ | Partial | — | — | ✓ | ✓ |
| Fleet tracking built-in | ✓ | Partial | — | — | ✓ | ✓ |
| No per-request pricing | ✓ | — | — | ✓ | — | — |
| Custom map styles | ✓ | Partial | ✓ | ✓ | Partial | Partial |
Koder Map uses OpenStreetMap as its primary data source, enriched with open datasets for elevation, satellite imagery, and address databases. You can also import custom data in GeoJSON, Shapefile, or MBTiles format.
Yes. Download map regions as compact vector tile packages for fully offline maps, geocoding, and routing. A country-level offline pack is typically 100-500 MB depending on detail level.
Koder Map uses OSRM and Valhalla routing engines with traffic-aware profiles. For areas with good OSM coverage, routing quality is comparable. You can integrate live traffic feeds to improve ETA accuracy.
No. Koder Map is self-hosted with no per-request fees. You can serve unlimited map tiles, geocoding requests, and routing queries from your own infrastructure.
Absolutely. Koder Map supports a flexible style specification similar to Mapbox GL Style Spec. Create branded map styles, dark mode maps, or data visualization themes with full control over colors, fonts, icons, and layers.
Maps, Navigation & Location Intelligence Platform