AutoGluon

AutoGluon is a Python AutoML library for developers and data scientists who want predictive models without selecting every algorithm themselves. It chooses combinations of classic machine learning and foundation models for the task.

• Table predictions: Predict outcomes from structured data, including tables with text and images, then compare model results and inspect feature importance.
• Time series forecasting: Forecast future values with Chronos-2 and ensemble models.
• Text, image and document models: Build classifiers for text, images, scanned documents and PDFs, or train models for object detection, semantic segmentation and named entity recognition.
• Search and matching: Extract embeddings for semantic search and compare text or images, including image-to-text matching; CLIP also supports image classification without task-specific training examples.
• Custom models and evaluation: Add your own models, feature processing and metrics when the built-in choices don't fit your problem.

AutoGluon is open source under Apache 2.0 and runs locally on Linux, macOS and Windows. It supports GPUs and parameter-efficient fine-tuning, with TensorRT optimization for faster prediction. Deployment options include pre-built containers and cloud predictors; cloud deployments run on hosted infrastructure.