# AutoGluon Assistant

A multi-agent system that automates end-to-end multimodal machine learning workflows, turning raw data into a trained ML solution with no manual pipeline building, across CLI, Python API, web UI and MCP interfaces.

AutoGluon Assistant, also called MLZero, is a multi-agent system that automates end-to-end multimodal machine learning and deep learning workflows, transforming raw multimodal data into a working ML solution without manual intervention.

It supports multiple LLM providers, with AWS Bedrock as the default alongside OpenAI, Anthropic and SageMaker, and can be used from a CLI, a Python API, a web UI or an MCP interface. A chat mode answers machine-learning questions conversationally without running any code. A node-based manager combined with Monte Carlo tree search improves on earlier greedy iterative-refinement approaches to searching over candidate solutions.

It currently runs on Linux only, inside a conda environment.

Website: https://github.com/autogluon/autogluon-assistant