https://github.com/ppmpreetham/mindmesh
two AI's interact with each other, using an initial prompt
https://github.com/ppmpreetham/mindmesh
ai ai-interaction
Last synced: 2 months ago
JSON representation
two AI's interact with each other, using an initial prompt
- Host: GitHub
- URL: https://github.com/ppmpreetham/mindmesh
- Owner: ppmpreetham
- License: mit
- Created: 2024-09-14T15:37:30.000Z (9 months ago)
- Default Branch: main
- Last Pushed: 2024-11-12T05:16:03.000Z (7 months ago)
- Last Synced: 2025-02-01T15:46:03.299Z (4 months ago)
- Topics: ai, ai-interaction
- Language: Python
- Homepage:
- Size: 636 KB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Mindmesh
An experimental project where multiple AI agents engage in self-dialogue, transforming an initial input prompt into something more complex / insightful.
## Features
*Multi-AI Collaboration*: Multiple AI agents engage in self-initiated conversations.
*Iterative Prompt Enhancement*: System transforms initial prompts into richer outputs over time.
*Research-Oriented*: Provides insights into AI-based creativity and conversational dynamics.
## Installation
To run this project locally, clone the repository and install the dependencies:
- [Ollama](www.ollama.com/)
- Choose a model of your choice. `It's Recommended to use llama2-uncensored`
```bash
git clone https://github.com/yourusername/mindmesh.git
cd mindmesh
```
Usage
- run `ollama serve` on both the server and client devices
- on one device, run the `server.py` and `client.py` in the other
Start by providing an initial prompt.
The AIs will engage in conversation, building on each other’s ideas.
Watch the output evolve as the conversation progresses.
Analyze the final result, comparing it to the initial input to observe the changes and improvements.
Contribution
This project is a work in progress, and contributions are welcome! Whether it's improving the AI models, adding new features, or enhancing the documentation, feel free to create a pull request or submit an issue.