Node management in CLI

Note: Users can use their browser and go to localhost:11025/docs to view the node API and see node stats and information. Models can also be managed under the node API without having to use a terminal curl request.

Here’s the full list of CLI commands to manage your Infera nodes.

Get node stats:

curl -X 'GET' \
 'http://localhost:11025/node_stats' \
 -H 'accept: application/json'

Get points:

curl -X 'GET' \
 'http://localhost:11025/points' \
 -H 'accept: application/json'

Get installed models:

curl -X 'GET' \
 'http://localhost:11025/installed_models' \
 -H 'accept: application/json'

Install models:

While the node is running, you can install llama3:latest (or another AI model listed below) using the following terminal command in a second terminal window:

curl -X POST http://localhost:11025/install_new_model \
     -H "Content-Type: application/json" \
     -d '{"install_model_name": "llama3:latest"}'

The node should now have llama3:latest installed and your node is now fully operational on the network.

llama3:latest can also be replace with the following models

gemma:latest
gemma2:latest
dolphin-mistral:latest
mistral:latest
llama3:latest
llama3.1:latest
llama2-uncensored:latest

Node details:

curl -X 'GET' \
 'http://localhost:11025/node_details' \
 -H 'accept: application/json'

Register node:

curl -X 'GET' \
 'http://localhost:11025/register_node' \
 -H 'accept: application/json'

Deregister node:

curl -X 'GET' \
 'http://localhost:11025/deregister_node' \
 -H 'accept: application/json'

Ping:

curl -X 'GET' \
 'http://localhost:11025/ping' \
 -H 'accept: application/json'

For any feedback and/or assistance please contact the team in our Discord. Please contact us in general chat or open a ticket.