Skip to content

Conversation

@pr0m1th3as
Copy link
Contributor

Hi, this PR add an additional public method in ollama class to facilitate memory management in the Ollama server. By default, models are removed from memory after 5 minutes of inactivity, but an application utilizing a number of different models might benefit from an unload method.

I have already implemented this method in my own version of your library I use for the llms Octave Package and it works great. However, I have not tested this with the CI of your repository, despite adding the method in the test suite.

I did not commit a changeset in the README file because I am not sure whether you prefer to group this method with the load method or have it in a separate sub-section.

This method facilitates memory management in the Ollama server.
Updated test case name to reflect loading and unloading models.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant