# LGM — Large Geospatial Model

> LGM (Large Geospatial Model) is a backend geospatial intelligence platform that combines raster and vector data, spatial computation, GeoRAG and AI orchestration. It helps people and applications investigate territories and return evidence, indicators and map-ready results.

LGM is the product name; Synta LGM is the identity used on synta.global. The platform is jointly developed by SynPlanet and GUIDE Digital. LGM is not simply one language model: specialized models are one part of its development roadmap.

This is a curated guide to public HTML pages. Product availability is staged: QGIS is described as private beta; specialized model-provider endpoints, Computer Vision and additional connectors are roadmap directions. Check the linked product pages and account configuration for current scope. No private collections, credentials or internal documents are included.

## Start here

- [LGM overview](https://synta.global/overview): Product definition, interfaces, availability, data limitations and next steps.
- [LGM website](https://synta.global/): Main product introduction and entry points.
- [About LGM](https://synta.global/company): Development partnership between SynPlanet and GUIDE Digital.

## Products and connections

- [Web GIS & Chat](https://synta.global/gis): Browser. Investigate a territory on a map with LGM chat. The application at gis.synta.global brings together Mapbox maps, Cesium 3D and Google Earth content.
- [Platform Dashboard](https://synta.global/platform): Account. The account surface for registration, sign-in, API keys, uploaded layers and billing. Available controls depend on the rollout and account configuration.
- [MCP Server](https://synta.global/mcp): AI agents. Expose LGM data and spatial tools to compatible agents through an authenticated remote MCP connection. Use the endpoint and credentials issued for your account.
- [API](https://synta.global/api): Applications. Integrate geospatial data and computation into your own application. API keys identify access and usage; supported operations depend on the service.
- [QGIS Plugin](https://synta.global/qgis): Private beta. Bring LGM layers and analysis into desktop GIS. The current public site describes private-beta access, not an unrestricted marketplace release.
- [Model Provider](https://synta.global/model-provider): Roadmap. Future specialized and fine-tuned geospatial models for partner catalogs. This is a development direction, not a claim of currently published model endpoints.
- [Geospatial RAG, Tools & Skills](https://synta.global/geospatial-rag-tools-skills): Spatial tools, geographic evidence retrieval and reusable workflows.
- [Orchestration & Agent Harnesses](https://synta.global/orchestration-agent-harnesses): Workflow coordination and scoped implementation of agent systems.
- [Developer guide](https://synta.global/developers): Choose a connection path and follow integration guidance.
- [Integration directory](https://synta.global/integrations): Current access paths and planned GIS, automation and digital-twin integrations.

## Data and commercial information

- [Data and formats](https://synta.global/data): Raster/vector collections, COG, GeoParquet, GeoJSON and STAC; source coverage and availability vary.
- [Pricing](https://synta.global/pricing): Usage-based model and enterprise options; check published rates and account terms rather than assuming a free allowance.

## Solutions

- [LGM solutions](https://synta.global/solutions): Territorial research, city planning, site selection, environment and infrastructure use cases.
- [Geospatial Data Layers](https://synta.global/solutions/geospatial-data-layers): Regional open, premium and private data preparation, storage and delivery engagements.
- [Geospatial & City Planning Benchmark](https://synta.global/solutions/geospatial-city-planning-benchmark): Scoped evaluation and improvement work; not a published model leaderboard.
- [AI Solutions for Partner Clients](https://synta.global/solutions/partner-client-solutions): Custom pilot and production implementation with partner models and LGM services.

## Research and publications

- [Tbilisi research](https://synta.global/research/tbilisi): Exploratory city screening, evidence and limitations; not approved construction or investment decisions.
- [Publications](https://synta.global/publications): Index of public articles.
- [How to evaluate geospatial AI for city planning](https://synta.global/publications/benchmarking-geospatial-ai-city-planning): A practical framework for testing spatial reasoning, data selection, tool calls and GIS-ready outputs, and turning failures into an improvement roadmap.
- [Geospatial agents need more than a language model](https://synta.global/publications/geospatial-agent-orchestration-and-harnesses): Understand how orchestration and agent harnesses connect planning, data retrieval, spatial tools and verification in a geospatial AI workflow.
- [From a 3D city to a decision: geospatial AI for digital twins](https://synta.global/publications/geospatial-intelligence-for-digital-twins): Explore how LGM can complement digital twins with territorial context, spatial analysis and traceable results without replacing the existing platform.
- [Geospatial site selection: from a business question to a shortlist](https://synta.global/publications/geospatial-site-selection-from-question-to-shortlist): Explore how LGM can structure site selection around demand, accessibility, competition and constraints, with evidence that explains every candidate.
- [Private geospatial data: bringing local knowledge into LGM](https://synta.global/publications/private-geospatial-data-shared-context): See how project uploads, shared collections and public LGM layers can work together, and what to clarify before using private data in spatial analysis.
- [What is LGM? A guide to the Large Geospatial Model](https://synta.global/publications/what-is-large-geospatial-model): Learn how LGM connects geospatial data, spatial computation and AI to turn territorial questions into evidence, map-ready layers and practical decisions.
- [Geospatial RAG: connecting AI answers to real places](https://synta.global/publications/geospatial-rag-grounded-spatial-answers): Discover how LGM uses Geospatial RAG to combine spatial context, data retrieval and GIS tools for answers grounded in a specific territory.
- [Connecting geospatial AI through MCP and API](https://synta.global/publications/connect-geospatial-ai-mcp-api): See how LGM fits into AI-agent and application workflows through MCP and API, from choosing a study area to returning usable geospatial results.
- [Raster and vector data: the foundation of geospatial AI](https://synta.global/publications/raster-vector-data-layers-for-ai): Understand how LGM combines raster and vector layers, cloud-oriented formats and private collections to support meaningful geospatial analysis.
- [Tbilisi: turning a citywide question into spatial priorities](https://synta.global/publications/tbilisi-geospatial-ai-research): Explore what LGM’s Tbilisi investigation revealed about urban access, infrastructure and site selection—and how to turn screening into local validation.

## Product applications

- [LGM GIS application](https://gis.synta.global/): Map and LGM chat application; some actions may require account access.
- [LGM Platform application](https://platform.synta.global/): Registration, account access, keys, collections and billing; private content requires authorization.