app.py

import subprocess
import os
from llama_cpp import Llama
import gradio as gr

# 🔹 Get Hugging Face Token from environment variable
HF_TOKEN = os.getenv("HF_TOKEN")

# 🔹 Model details
MODEL_URL = "https://huggingface.co/TheBloke/Mistral-7B-GGUF/resolve/main/mistral-7b.Q4_K_M.gguf"
MODEL_PATH = "./models/mistral-7b.Q4_K_M.gguf"

# 🔹 Ensure the models directory exists
os.makedirs("./models", exist_ok=True)

# 🔹 Check if the model exists, else download it
if not os.path.exists(MODEL_PATH):
    print("🚀 Downloading Mistral-7B Q4 GGUF model...")
    subprocess.run([
        "wget", "--header", f"Authorization: Bearer {HF_TOKEN}",
        MODEL_URL, "-O", MODEL_PATH
    ], check=True)
    print("✅ Download complete!")

# 🔹 Load the model
print("📥 Loading the model...")
model = Llama(model_path=MODEL_PATH, n_ctx=4096, n_threads=8)
print("✅ Model loaded successfully!")

# 🔹 Define a function to interact with the model
def chat_with_mistral(prompt):
    response = model(prompt, max_tokens=512)
    return response["choices"][0]["text"]

# 🔹 Create a Gradio UI
iface = gr.Interface(fn=chat_with_mistral, inputs="text", outputs="text", title="Mistral-7B Chatbot")

# 🔹 Launch the app
iface.launch()