Metadata-Version: 2.3
Name: llm-snowglobe
Version: 0.5.0
Summary: Snow Globe multi-agent system for open-ended wargames with large language models
License: Apache-2.0
Author: Daniel Hogan
Author-email: 6313241+dphogan@users.noreply.github.com
Requires-Python: >=3.10
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Requires-Dist: accelerate
Requires-Dist: fastapi
Requires-Dist: jupyter
Requires-Dist: langchain
Requires-Dist: langchain-chroma
Requires-Dist: langchain-community
Requires-Dist: langchain-huggingface
Requires-Dist: langchain-openai
Requires-Dist: langchain-text-splitters
Requires-Dist: llama-cpp-python (>=0.3.8)
Requires-Dist: markdown2
Requires-Dist: nicegui
Requires-Dist: numpy
Requires-Dist: openai
Requires-Dist: platformdirs
Requires-Dist: pydantic
Requires-Dist: pyyaml
Requires-Dist: torch
Requires-Dist: torchvision
Requires-Dist: tqdm
Requires-Dist: transformers
Requires-Dist: twine
Requires-Dist: uvicorn
Requires-Dist: watchfiles
Project-URL: Homepage, https://github.com/IQTLabs/snowglobe
Description-Content-Type: text/markdown

# Snow Globe
***Open-Ended Wargames with Large Language Models***

Snow Globe, an ongoing applied research project and resulting software package, uses large language models (LLMs) for automated play of "open-ended" text-based wargames, such as seminar games and political wargames.  LLMs enable a light, flexible architecture in which player actions are not restricted to predefined options.  The system allows humans to play against or alongside AI agents with specific personas.  Every stage of the wargame from scenario preparation to post-game analysis can be optionally carried out by AI, humans, or a combination thereof.

Read more [here](https://arxiv.org/abs/2404.11446).

## Installation

Build the Docker image and run a container.

```
./docker_setup.sh
```

Or, install Snow Globe from PyPI.  For CPU only:

```
pip install llm-snowglobe
```

For GPU support:

```
CMAKE_ARGS="-DGGML_CUDA=on" pip install llm-snowglobe
```

## Demos

After installation, you can simulate a tabletop exercise about an AI incident response.

```
examples/haiwire.py
```

Or, simulate a political wargame about a geopolitical crisis.

```
examples/ac.py
```

In the latter case, you can use a terminal-based chat interface to discuss the game afterwards, or just press `Enter` twice to exit.

## Human Players

To play a game between a human and an AI player, launch the server and start a game:

```
snowglobe_server &
examples/ac.py --human 1
```

Then, open a browser window and navigate to:

```
http://localhost:8000
```

The terminal output will include the ID number for the human player.  Type the number into the ID box, click `Log In`, then click `Chat` to enter the human player's responses using a graphical user interface.

Make sure to run `snowglobe_server` from the same file system location where you run the game.  Files related to the graphical user interface will be stored in that location.

## License

This repo is released under the [Apache License Version 2.0](LICENSE), except for [jQuery](src/llm_snowglobe/terminal/jquery-3.7.1.min.js) which is released under the [MIT License](https://github.com/jquery/jquery/blob/main/LICENSE.txt).

