Open Source Machine Learning Infrastructure Software
Machine Learning Infrastructure software provides the underlying systems required to train, deploy, and manage artificial intelligence models. These tools handle the orchestration of GPU clusters, the storage of high-dimensional vectors for similarity search, and the tracking of training metrics. This type of software is for data scientists and machine learning engineers who need to manage large datasets and model parameters. Self-hosting these apps ensures that proprietary training sets and model weights remain on private hardware instead of on external servers.
This page lists 7 open source tools in the Machine Learning Infrastructure category. The most popular are Milvus, Netron and Databend. Most use the Apache-2.0 or AGPL-3.0 license, and 5 offer an official Docker image.
Milvus
A distributed vector database for organizing and searching unstructured data through similarity search and metadata filtering.
Netron
Visualize neural network, deep learning and machine learning models through a comprehensive graphical interface.
Databend
An open-source cloud data warehouse for large-scale analytics, vector search, and full-text search using Rust.
Transformer Lab
A unified environment to train, evaluate, and scale large language and diffusion models on local or cluster hardware.
Deepnote
A collaborative data notebook for analysts and scientists that integrates Python, SQL, and AI assistance.
dstack
A unified control plane for GPU provisioning and container orchestration across cloud and on-premise clusters.
mlop
A self hosted framework for machine learning engineers to track metrics, parameters, and gradients during model training.
Join our newsletter to get shiny new open source software delivered to your inbox. Unsubscribe anytime.