AgentNet is a sophisticated peer-to-peer networking framework that empowers distributed agents to communicate and share data with high efficiency. Leveraging Vulkan for graphics rendering, AgentNet is built for performance and scalability, providing a robust platform for decentralized agent interactions.
These instructions will guide you through getting a copy of the project up and running on your local machine for development and testing purposes.
Before you begin, ensure you have the following installed:
- Vulkan SDK
- CMake
- A C++ compiler like GCC or Clang
- Access to Llama sources (see below)
To build this project from source, you will need access to the Llama sources. Follow these steps to download them:
- Visit the Llama GitHub repository and follow the instructions to register and accept the licenseĀ¹.
- Once approved, you will receive a custom URL via email from Meta.
- Clone the Llama repository and run the
download.sh
script with the custom URL provided. (Requires Linux or MacOS)
Follow these steps to set up your development environment:
- Clone the repository:
git clone https://github.com/K-Rawson/AgentNet.git
- Navigate to the project directory:
cd AgentNet
- Create a build directory:
mkdir build && cd build
- Run CMake to configure the project:
cmake ..
- Build the project:
cmake --build .
To run the automated tests for this system, use the following command:
ctest
For deployment, additional steps may be required, such as setting up a server environment or configuring network settings.
- Vulkan - A low-overhead, cross-platform 3D graphics and compute API.
- CMake - An open-source, cross-platform family of tools designed to build, test, and package software.
- Llama - For building AgentNet llama models from source. (Advanced)
Please read CONTRIBUTING.md for details on our code of conduct, and the process for submitting pull requests to us.
We use SemVer for versioning. For the versions available, see the tags on this repository.
- K-Rawson - Initial work - AgentNet
See also the list of contributors who participated in this project.
This project is licensed under the MIT License - see the LICENSE.md file for details.
- Tonic-AI
Please cite this project as:
Rawson, K. (2024). AgentNet: An Innovative AI Framework for Gaming. [https://github.com/K-Rawson/AgentNet](https://github.com/K-Rawson/AgentNet)