🧠 Local Mistral LLM on MacBook (Apple Silicon)

Run the powerful open-source Mistral 7B Instruct language model completely offline on your MacBook Pro using Metal GPU acceleration.

No APIs. No token limits. No cost. Just raw, local intelligence 💻⚡

---

🚀 Features

• 🐍 Python interface using llama-cpp-python
• ⚡ Optimized for Apple Silicon (M1, M2, M3, M4)
• 🧠 Run quantized Mistral 7B (4-bit) models locally
• 📝 Easily modify prompts, automate tasks, and build cool stuff

---

📦 Requirements

• macOS with Apple Silicon (M1 or newer)
• Python 3.10+
• Metal GPU support (comes with macOS)
• llama-cpp-python

---

🛠️ Setup Instructions

1. Clone the repo and create your project folder

mkdir -p ~/Documents/python-projects/mistral-demo
cd ~/Documents/python-projects/mistral-demo
python3 -m venv venv
source venv/bin/activate

2. Install dependencies

pip install -r requirements.txt

3. Download the model

mkdir -p ~/mistral-model
cd ~/mistral-model

curl -L -o mistral-7b.Q4_K_M.gguf \
https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.1-GGUF/resolve/main/mistral-7b-instruct-v0.1.Q4_K_M.gguf

(Optional)

Generate Custom Regional Language Dataset Using Google Translate API

1. Activate Google Translate API (First $300 USD Free)

2. Install google-cloud-translate

    pip install google-cloud-translate

3. Run the script

python3 google_translate.py

Tips

Keep your MacBook awake while running scripts

caffeinate -di python your_script.py