Mundi.ai is an open-source, AI-native web GIS designed for hosting, exploring, and analyzing diverse spatial data. It targets engineers, researchers, and power users seeking advanced spatial data capabilities, offering a benefit of integrating AI, particularly LLMs, directly into GIS workflows for tasks like geoprocessing and symbology editing. The project provides both a free cloud-hosted trial and a self-hostable open-source version.

How It Works

Mundi.ai employs an AI-native architecture, supporting vector, raster, and point cloud data formats. It seamlessly connects to and queries spatial databases such as PostGIS. A key innovation is its use of Large Language Models (LLMs) to interpret natural language commands for executing geoprocessing algorithms and dynamically editing map symbology, offering a novel, intuitive interface for complex spatial operations.

Quick Start & Requirements

A free trial is available on Mundi cloud at app.mundi.ai. For self-hosting, users require git, Docker, and a capable computer or server. Optional integration with local LLMs or any provider supporting the chat completions API is available. A self-hosting tutorial is provided.

Highlighted Details

• Supports vector, raster, and point cloud data.
• Integrates with PostGIS for data querying and layer management.
• Leverages LLMs for geoprocessing and symbology control.
• Offers self-hosting capabilities with local LLM support.
• Optional integration with QGIS for geoprocessing tasks.

Maintenance & Community

Contributions are welcomed via GitHub issues and Discord. A Contributor License Agreement (CLA) is required. Discussions on potential contributions are encouraged on Discord or GitHub. Security concerns should be reported directly to support@buntinglabs.com.

Licensing & Compatibility

The core Mundi.ai project is licensed under AGPLv3. Components that link with QGIS for geoprocessing are licensed under GPLv3. The AGPLv3 license imposes strong copyleft requirements, which may impact integration with closed-source applications.

Limitations & Caveats

Self-hosting necessitates a robust hardware setup. Contributions introducing significant new dependencies will be evaluated against the overhead they add to self-hosting. The AGPLv3 license may present compatibility challenges for certain commercial or closed-source use cases.