How to clear the chat context?

[N8python]

I want to load the context of a model once, query it with a prompt, and then be able to query it again without the model remembering the first prompt. Is this doable with the current library API?

[giladgd]

@N8python This is not possible ATM, but it'll be possible as part of #85 that I'm working on right now

[N8python]

Awesome! Is there a workaround I could use in the meantime, bar loading the entire Llama context again?

[giladgd]

@N8python If this is still relevant, see #125 for a workaround

[N8python]

Thank you!