Skip to content

Installation

Agentomics-ML supports multiple deployment options. Choose the one that best fits your needs.

Prerequisites

All installation methods require:

git clone https://github.com/BioGeMT/Agentomics-ML.git
cd Agentomics-ML

Docker with Pre-built Images

Fastest setup - Downloads pre-built images from Docker Hub.

Requirements

  • Docker installed and running

Setup

# Create a .env file (required for Docker mode)
cp .env.example .env
# Edit .env and set at least one API key

# Run with pre-built images
./run.sh --pull-images

The images will be downloaded automatically on first run.

Docker with Local Build

Default mode - Builds Docker images locally.

Requirements

  • Docker installed and running

Setup

# Create a .env file (required for Docker mode)
cp .env.example .env
# Edit .env and set at least one API key

# Run (will prompt to build images on first run)
./run.sh

On first run, you'll be prompted to build the Docker images. This takes a few minutes but only needs to be done once.

Local Mode (No Docker)

For development or Google Colab - Runs directly with conda.

Security Notice

Local mode executes code without containerization. Only use in secure environments like Google Colab or your own isolated container.

Requirements

Setup

# Set your API key (export or .env)
export OPENROUTER_API_KEY="your-key-here"

# Run in local mode
./run.sh --local

Conda environments will be created automatically.

Google Colab

The easiest way to try Agentomics-ML without any local setup.

Open in Google Colab

The Colab notebook uses local mode automatically.

Ollama (Local LLMs)

Run with local models using Ollama for privacy or offline use.

Requirements

  • Ollama installed and running
  • Docker (recommended) or conda

Docker Mode Setup

  1. Ensure Ollama listens on the host (e.g., 0.0.0.0:11434).

  2. Run with the --ollama flag:

    ./run.sh --ollama

Docker mode connects to the URL configured in src/utils/providers/configured_providers.yaml (default: http://host.docker.internal:11434/v1).

Local Mode Setup

For local mode, set the Ollama base URL in src/utils/providers/configured_providers.yaml to http://localhost:11434/v1, then run:

./run.sh --local

CPU-Only Mode

Disable GPU acceleration:

./run.sh --cpu-only

Works with both Docker and local modes.

Comparison Table

Mode Docker Required Build Time Security Best For
Docker + Pull Images Yes None High Quick start
Docker + Local Build Yes ~5-10 min High Custom builds
Local Mode No ~2 min Low Development, Colab
Google Colab No None Medium Trying it out
Ollama Depends Varies High Privacy, offline

Next Steps