Llama3 8B 1.58 100B Tokens GGUF

🚀 Llama3-8B-1.58 Models

This project focuses on the Llama3-8B-1.58 models. It converts the model to GGUF format from HF1BitLLM/Llama3-8B-1.58-100B-tokens using llama.cpp, providing users with a convenient way to use these models.

📦 Installation

Install llama.cpp via Brew

You can install llama.cpp through Brew, which works on both Mac and Linux systems.

brew install llama.cpp

💻 Usage Examples

Basic Usage

Use CLI

llama-cli --hf-repo brunopio/Llama3-8B-1.58-100B-tokens-GGUF --hf-file Llama3-8B-1.58-100B-tokens-GGUF -p "The meaning to life and the universe is"

Use Server

llama-server --hf-repo brunopio/Llama3-8B-1.58-100B-tokens-GGUF --hf-file Llama3-8B-1.58-100B-tokens-GGUF -c 2048

Advanced Usage

You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo.

Step 1: Clone llama.cpp from GitHub

git clone https://github.com/ggerganov/llama.cpp

Step 2: Build llama.cpp

Move into the llama.cpp folder and build it with the LLAMA_CURL=1 flag along with other hardware - specific flags (e.g., LLAMA_CUDA=1 for Nvidia GPUs on Linux).

cd llama.cpp && LLAMA_CURL=1 make

Step 3: Run Inference

Run inference through the main binary.

./llama-cli --hf-repo brunopio/Llama3-8B-1.58-100B-tokens-GGUF --hf-file Llama3-8B-1.58-100B-tokens-GGUF -p "The meaning to life and the universe is"

or

./llama-server --hf-repo brunopio/Llama3-8B-1.58-100B-tokens-GGUF --hf-file Llama3-8B-1.58-100B-tokens-GGUF -c 2048

📚 Documentation

Model Information