A geospatial AI toolbox that provides tools for training and using machine learning models with geospatial and tabular data, featuring automated ML for classification and regression, plus NLP capabilities for text analysis.
ArcGIS Pro GeoAI Toolbox is a Geospatial AI toolset built into Esri's ArcGIS Pro that enables users to train, fine-tune, and apply machine learning and deep learning models on geospatial, imagery, tabular, text, and time-series data, with pricing bundled into ArcGIS Pro licensing starting at $700/year for a Basic license. It targets GIS analysts, data scientists, and spatial researchers working within the Esri ecosystem.
The toolbox is organized into four distinct toolsets, each addressing a major class of geospatial AI problems. The Feature and Tabular Analysis toolset leverages automated machine learning (AutoML) to train, fine-tune, and ensemble the best-performing models for classification and regression on feature classes and tabular datasets. The Imagery AI toolset applies object detection and pixel classification deep learning algorithms to raster and imagery data, powering workflows like building footprint extraction, land cover classification, and damage assessment. The Text Analysis toolset uses natural language processing to classify documents, transform text, and extract named entities such as addresses, while the Time Series AI toolset produces forecasts and future value estimates across space-time cubes.
Unlike standalone ML platforms, GeoAI Toolbox is tightly integrated with ArcGIS Living Atlas of the World, which provides a library of pretrained geospatial models that can be used directly or fine-tuned on local data. Models created in the toolbox interoperate with the ArcGIS API for Python's arcgis.learn module, allowing advanced users to extend and refine workflows in Python and Jupyter notebooks. Based on our analysis of 870+ AI tools, GeoAI Toolbox stands out among Geospatial AI offerings because it embeds deep learning directly into a production GIS platform used by more than 350,000 organizations worldwide, rather than requiring users to stitch together separate ML and mapping stacks. Compared to open-source alternatives like QGIS with plugins or Google Earth Engine, it offers a more unified, enterprise-ready environment, though it requires a paid ArcGIS Pro license and installation of the Deep Learning Libraries Installer for full functionality.
Was this helpful?
The Feature and Tabular Analysis toolset uses AutoML to train, fine-tune, and ensemble the best models for classification and regression on feature classes and tables. Users provide labeled data and the tool evaluates multiple algorithms against available compute resources to select an optimal pipeline. The resulting model can predict categorical or continuous target variables on similar datasets.
The Imagery AI toolset applies deep learning models to imagery for tasks like building extraction, land cover classification, and object counting. It can consume pretrained models from ArcGIS Living Atlas or custom-trained models exported from the arcgis.learn Python module. Outputs integrate directly with ArcGIS Pro raster and feature workflows.
The Text Analysis toolset performs classification, transformation, and named-entity extraction on unstructured text, including geographic entities like addresses. It supports pretrained NLP models from Living Atlas as well as user-trained models built from labeled text data. This lets organizations unlock insights from reports, permits, and other text-heavy records and join them back to spatial data.
The Time Series AI toolset forecasts and estimates future values at specific locations within a space-time cube, supporting applications like demand planning, environmental monitoring, and risk assessment. It combines temporal patterns with spatial structure to produce location-aware forecasts rather than purely temporal ones. Results are returned as cubes or feature layers ready for mapping and further analysis.
GeoAI Toolbox tools can directly consume pretrained models from ArcGIS Living Atlas of the World, a curated library of geospatial AI models maintained by Esri. Models created or fine-tuned by the toolbox are fully interoperable with the arcgis.learn module of the ArcGIS API for Python. This allows advanced users to script, extend, and deploy models outside the ArcGIS Pro UI.
$700/year
$2,500/year
$2,700/year
Ready to get started with ArcGIS Pro GeoAI Toolbox?
View Pricing Options âWe believe in transparent reviews. Here's what ArcGIS Pro GeoAI Toolbox doesn't handle well:
Weekly insights on the latest AI tools, features, and trends delivered to your inbox.
ArcGIS Pro 3.4 (released late 2024) and 3.5 (early 2025) expanded the GeoAI Toolbox with new tools for point cloud classification using deep learning, additional pretrained foundation models in Living Atlas, and improved support for transformer-based architectures in the Imagery AI toolset. The Deep Learning Libraries Installer was updated to support PyTorch 2.x and newer CUDA versions, simplifying GPU setup. Esri also introduced GeoAI-powered anomaly detection tools and enhanced the Time Series AI toolset with additional forecasting methods. The 2025 Living Atlas update added new pretrained models for global building footprint extraction and land use classification trained on recent high-resolution imagery.
No reviews yet. Be the first to share your experience!
Get started with ArcGIS Pro GeoAI Toolbox and see if it's the right fit for your needs.
Get Started âTake our 60-second quiz to get personalized tool recommendations
Find Your Perfect AI Stack âExplore 20 ready-to-deploy AI agent templates for sales, support, dev, research, and operations.
Browse Agent Templates â