LLM Party Chat
A real-time chat system that enables multiple Large Language Models to engage in conversations with each other and human moderators. Models can run on different machines and communicate through a central websocket server.
Features
- Multi-model conversation support
- Real-time websocket communication
- Human moderation interface
- Color-coded messages for different participants
- Support for any Hugging Face transformers model
- Distributed architecture - models can run on different machines
- Graceful handling of connections/disconnections
Installation
Method 1: Quick Setup (Recommended for trying it out)
- Clone the repository:
git clone https://github.com/yourusername/llm-party-chat.git
cd llm-party-chat
- Create and activate a virtual environment:
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
- Install dependencies:
pip install websockets transformers torch colorama aioconsole
Method 2: Install as Package
pip install llm-party-chat
Usage
Method 1: Direct Usage (Recommended for development)
- Start the server:
python server.py
- In separate terminals, start two or more model clients:
python client.py --name "Model1"
python client.py --name "Model2"
- Start the moderator interface:
python moderator.py
Method 2: Package Usage
If you installed via pip:
- Start the server:
python -m llm_party_chat.server
- In separate terminals, start model clients:
python -m llm_party_chat.client --name "Model1"
python -m llm_party_chat.client --name "Model2"
- Start the moderator:
python -m llm_party_chat.moderator
Components
Server (server.py)
- Central websocket server that manages connections
- Handles message broadcasting
- Manages client registration/disconnection
- Maintains chat history
- Color codes different participants
Client (client.py)
- Loads and runs a language model
- Connects to the server
- Processes incoming messages
- Generates responses using the model
- Supports various model configurations
Moderator (moderator.py)
- Human interface to the chat
- Sends prompts to models
- Monitors all conversations
- Views system status and connections
Configuration
Client Configuration
python client.py \
--name "Model1" \
--model "TinyLlama/TinyLlama-1.1B-Chat-v1.0" \
--max-tokens 50 \
--temperature 0.7 \
--server "ws://localhost:8765"
Server Configuration
python server.py --host "localhost" --port 8765
Requirements
- Python 3.10+
- websockets
- transformers
- torch
- colorama
- aioconsole
Development
The repository structure:
llm-party-chat/
├── LICENSE
├── MANIFEST.in
├── README.md
├── requirements.txt
├── setup.py
└── src/
└── llm_party_chat/
├── __init__.py
├── server.py
├── client.py
└── moderator.py
For development:
- Clone the repository
- Create a virtual environment
- Install requirements
- Run the components directly using Method 1 above
Future Improvements
- Message persistence
- Web interface
- More model options
- Chat history export
- Authentication
- Docker support
License
MIT License
Contributing
- Fork the repository
- Create a new branch (
git checkout -b feature/improvement) - Make changes
- Commit (
git commit -am 'Add feature') - Push (
git push origin feature/improvement) - Create Pull Request
Metadata
Release files for llm-party-chat 0.0.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| llm_party_chat-0.0.1.tar.gz | 8.9 kB | Without provenance | Details for llm_party_chat-0.0.1.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| llm_party_chat-0.0.1-py3-none-any.whl | Python 3 | none | any | Details for llm_party_chat-0.0.1-py3-none-any.whl |
Total release size: 17.9 kB
Release files / llm_party_chat-0.0.1.tar.gz
| Download URL | llm_party_chat-0.0.1.tar.gz |
|---|---|
| Size | 8.9 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
1c0c624c3096d2754f9ae3b25c0c954228f629a53bf97779f2a8cc3fb2be6f5e
|
|
BLAKE2b-256 checksum How to use checksums |
7e55b9225dd11aad0e0a7eb7f7f1f2a360690646bd1dd2e17fd2f92135cf803d
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.10.13
|
Release files / llm_party_chat-0.0.1-py3-none-any.whl
| Download URL | llm_party_chat-0.0.1-py3-none-any.whl |
|---|---|
| Size | 9.0 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
b5ef781eb351bdb996c3b20e1b1152dde133f870117b50232c7b5719f6745da0
|
|
BLAKE2b-256 checksum How to use checksums |
bb01b981bb2d2e22008fa3d75fc4d3c1eb984bf65a9c90039fb1caeeaf52ddb9
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.10.13
|
Security for llm-party-chat 0.0.1
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.