News: Our paper "MLOX: Open-Source MLOps for the Rest of Us" has been accepted by CAIN 2026

Deploy and manage ML/AI infrastructure on your own servers.

Slothfully simple.

Servers, Docker, Kubernetes, databases, workflows, model serving, data services, tracking, and monitoring, with your product at the center. Managed in one place and connected by design.

MLOX Sloth Logo

WHY TEAMS PICK MLOX

Your Product at the Center

Keep your product at the center.

MLOX manages servers, sets up Docker and Kubernetes clusters, deploys open-source services, stores project state and secrets in an encrypted project file, and wires dependencies across databases, workflows, experiment tracking, model serving, data services, and monitoring.

Deploy open-source components through the Web UI, TUI, or CLI. MLOX keeps the stack inspectable, keeps each project in one portable encrypted file, and connects service dependencies across your own servers and external integrations.

Built for solopreneurs, startups, and small teams that want to build their product instead of assembling infrastructure.

Manage Your Stack

Run on your own servers with inspectable configurations and explicit dependencies.

Connect the Pieces

Keep databases, workflows, model serving, tracking, and monitoring connected around your product.

Choose Your Interface

Use the Web UI, TUI, or CLI against the same shared project model.

UI + TUI + CLI

One Shared Project Model

Open Source

Transparent By Default

Connected

Product And Services Together

What can you do with MLOX?

Infrastructure

  • Add, remove, tag, and secure servers
  • Use Native, Docker, or Kubernetes execution
  • Create and connect single- or multi-node clusters

Services

  • Deploy, manage, and remove ML/AI services
  • Centralize secrets and configuration
  • Connect databases, workflows, tracking, model serving, and monitoring
  • Use MLflow, Airflow, LiteLLM, Ollama, PostgreSQL, Redis, and more

Product & Integrations

Put your product in the same connected topology as the services around it:

  • Import public or private GitHub repositories
  • Deploy Docker Compose applications from managed repositories
  • Connect BigQuery, Secret Manager, Cloud Storage, Sheets, and more
  • Access infrastructure, secrets, and services from Python

Why Choose MLOX?

Open Source & Community Driven

Fully open-source under the MIT license, built by the community for the community. We believe in transparency and collaboration.

Own the Infrastructure

Run ML/AI services on your own servers and connect external providers where they make sense, without committing the whole stack to one platform.

Sloth-Paced Philosophy

We believe in thoughtful, calm systems over rushed chaos. Move fast by moving deliberately, with sensible defaults and clear documentation.

Connected by Design

Keep servers, services, secrets, dependencies, and your product in one portable encrypted project file and inspectable topology operated through the Web UI, TUI, or CLI.

See MLOX in Action

Manage the infrastructure around your product through the Web UI, TUI, and CLI.

MLOX Infrastructure Control

Infrastructure Control

Manage servers and set up Docker or Kubernetes

MLOX Service Management

Service Management

Deploy and connect ML/AI services

MLOX GIT & Cloud Integrations

Git & Cloud Integrations

Connect repositories, applications, and cloud services

MLOX Real-time Monitoring

Real-time Monitoring

Keep tracking and monitoring close to your stack

MLOX TUI

Terminal UI

Operate the same project model from the terminal

MLOX CLI

CLI

Script and automate infrastructure workflows

Get Started

Sloth-Friendly Setup

Easing into MLOX should feel like a lazy stretch on a sunny branch

  1. Install Task

    Install Task — our go-powered task runner

    # Visit https://taskfile.dev/installation/ for installation instructions

    Note: Task is a modern task runner / build tool that aims to be simpler and easier to use than GNU Make

  2. Clone the Repository

    Get the MLOX source code

    git clone https://github.com/BusySloths/mlox.git
  3. Run First Steps

    Mosey into the project and run the setup

    task first:steps

    Note: This unhurried command crafts a conda environment and gathers every dependency for you

  4. Create a Project

    Create one encrypted, portable MLOX project file

    mlox project new ./projects/demo

    Note: Set MLOX_PROJECT_PATH and MLOX_PROJECT_PASSWORD to open the same project from the Web UI, TUI, CLI, or SDK

  5. Want the Big Picture?

    Check out our comprehensive guides

    Note: For a more detailed guide, check out our Installation Guide. You can also view our MLOX – State of the Union — a short slide overview of what MLOX is, what problem it solves, and where it's heading.

Big Thanks to our Sponsors and Supporters

BMFTR PrototypeFund