---

LlamaCpp Server Installation Script

This shell script automates the installation, setup, and running of the LlamaCpp server along with the OpenChat 3.5-GGUF model. The script checks if LlamaCpp is already installed and if the required model is downloaded, then proceeds accordingly.

Requirements

• A Unix-based operating system (Linux, macOS, or WSL on Windows)
• Git
• Curl
• Make

Script Overview

Functions

• check_error(): Checks for errors after each command and exits if an error occurs.
• is_llama_installed(): Checks if the llama-server binary is already built.
• is_model_dowloaded(): Checks if the OpenChat 3.5-GGUF model is already downloaded.

Main Operations

1. Clone and Build LlamaCpp:

   • Clones the LlamaCpp repository from GitHub if it is not already installed.
   • Builds the llama-server binary using make.

2. Download the OpenChat Model:

   • Downloads the OpenChat 3.5-GGUF model from Hugging Face if it is not already present in the models directory.

3. Start the LlamaCpp Server:

   • Runs the LlamaCpp server with the specified model on port 8080.

Usage

1. Clone the Repository: Ensure you have the script saved in the desired directory.

2. Allowed permission

   chmod +x ./llamacpp.sh

3. Run the Script:

   ./llamacpp.sh

4. Script Execution:

   • The script will first check if llama-server is already installed.
   • If not, it will clone the llama.cpp repository and build the server.
   • Then, it checks if the OpenChat 3.5-GGUF model is already downloaded. If not, it will download the model.
   • Finally, it starts the llama-server using the downloaded model.

5. Access the Server: The server will be accessible at http://localhost:8080/.

Notes

• Ensure that all dependencies (Git, Curl, Make) are installed before running the script.
• The script assumes the presence of a Unix-like environment. Adjustments might be needed for other environments.

---