Configure Atoma Node

​

Overview

Atoma Node requires two configuration files:

• .env for inference services and PostgreSQL settings
• config.toml for node-specific configurations

​

Prerequisites

• Clone the Atoma Node repository

  git clone https://github.com/atoma-network/atoma-node
  

​

Configure .env

1. Copy the example file to create your .env:

cp .env.example .env

2. Update these required settings in your .env:

• POSTGRES_DB: Your PostgreSQL database name
• POSTGRES_USER: Your PostgreSQL username
• POSTGRES_PASSWORD: Your PostgreSQL password

3. Update the backend settings in your .env file based on the models and backends you plan to use. Ensure that each model or backend has the appropriate configuration settings specified.

​

Configure config.toml

1. Copy the example file to create your config.toml:

cp config.example.toml config.toml

2. Update these required settings in your config.toml:

• models: List the AI models your node will serve (e.g. ["meta-llama/Llama-3.2-3B-Instruct", "intfloat/multilingual-e5-large-instruct"])
• database_url: Set your PostgreSQL connection string using values from .env (e.g. "postgres://<POSTGRES_USER>:<POSTGRES_PASSWORD>@postgres-db:5432/<POSTGRES_DB>")
• node_small_ids: Set the list of small IDs for your node (e.g. [1, 2])
• node_badges: Set the list of node badges, where each badge is a tuple of (badge_id, small_id), both values are assigned once the node registers itself (e.g. ["0x268e6af9502dcdcaf514bb699c880b37fa1e8d339293bc4f331f2dde54180600",1])

Other configuration values can remain at their defaults.