llm-cache-plugin 0.1

Check whether you've already run this prompt before calling the LLM

llm-cache

Check whether you've already run this prompt before calling the LLM

Installation

Install this plugin in the same environment as LLM.

llm install llm-cache-plugin

Usage

This plugin add a --cache flag to llm prompt.

If --cache is specified, the plugin will search your llm database for a previously-saved response for this prompt-system-model combination. If it is found, it returns that text, saving you time and money. If not, it passes off to the regular llm prompt command.

llm 'How many rs are in strawberry? Think very hard.'
> There are 2 'R's in strawberry

Took 3.01s user 1.45s system 104% cpu 4.280 total

llm --cache 'How many rs are in strawberry? Think very hard.'
> There are 2 'R's in strawberry

Took 2.83s user 1.50s system 152% cpu 2.836 total

Development

To set up this plugin locally, first checkout the code. Then create a new virtual environment:

cd llm-cache
python -m venv venv
source venv/bin/activate

Now install the dependencies and test dependencies:

llm install -e '.[test]'

To run the tests:

pytest