Infernet CLI

Infernet CLI is a tool that simplifies configuration and deployment of an Infernet Node. Specifically, it enables:

1. Pulling plug-and-play node configurations for different chains, with the ability to further configure and customize them.
2. Adding plug-and-play service configurations to your node.
3. Creating, managing, and destroying a node.

Prerequisites

• Python >= 3.9
• Docker Desktop or (Docker Engine + Docker Compose).

Installation

You can either install infernet-cli via uv (recommended) or via pip.

uvpip

uv pip install infernet-cli

pip install infernet-cli

Quickstart

Here's how you can configure a node connected to a local Anvil chain:

export DEPLOY_DIR=deploy/

infernet-cli config anvil --skip

The output will look something like this:

No version specified. Using latest: v1.3.0
Using configurations:
   Chain = 'anvil'
   Version = '1.3.0'
   GPU support = disabled
   Output dir = 'deploy'

Stored base configurations to '/root/deploy'.
To configure services:
  - Use `infernet-cli add-service`
  - Or edit config.json directly

You can add an ML service, e.g. onnx-inference, as follows:

infernet-cli add-service onnx-inference --skip

The output will look something like this:

Version not provided. Using latest version '2.0.0'.
Successfully added service 'onnx-inference-2.0.0' to config.json.

You can then deploy the node:

infernet-cli start

and check that it's healthy:

infernet-cli health

More Options

To see all the available commands and options, head over to the Usage documentation.