LLMLabyrinth is a call back to the old school text based console adventure games with a twist. Descriptions are generated by a local GPT model in order to add dynamic elements to an exciting adventure.
Find and combine the 3 items to win the game!
- Directional walking (omg!)
- AI LLM generated room descriptions
- Room themes
- Basic ascii mapping system
- Inventory
- Item combinations
- Win state
-
Clone this repo
-
Set up a virtual enviroment
python3 -m venv env -
Activate the virtual environment
source env/bin/activate -
Install the requirements
pip3 install -r requirements.txt- install python dependencies -
Download the LLM model and place it in the
modelsdirectory- LLM: default to ggml-gpt4all-j-v1.3-groovy.bin. If you prefer a different GPT4All-J compatible model, just download it in the applicable folder
-
Run the game
python3 main.py
| Command | Action |
|---|---|
| n | north |
| e | east |
| s | south |
| w | west |
| m | map |
| i | check inventory |
| g | get items |
| c | combine items |
| q | quit |