Running Locally

Now that you have made your application and built it, you can run it locally to test it out.

To do so run the following command, where the -p is the path to the project.ts, and -h is the URL of the Ollama endpoint:

subql-ai -p ./path/to/project.ts -h  http://ollama.public.url

Once the project is running you should see the following: Listening on http://0.0.0.0:7827/

You can now interact with your application. The easiest way to do that is to run the repl in another terminal.

subql-ai repl

This will start a CLI chat. You can type /bye to exit.

Alternatively you can run with docker to get a web UI.

You may want to refer to the command line arguments used in SubQuery AI App Framework. It will help you understand the commands better.