app.py

import gradio as gr
from llama_cpp import Llama

MODEL_PATH = "model.gguf"  # downloaded in advance

llm = Llama(model_path=MODEL_PATH, n_ctx=8192, n_threads=4)

def chat(prompt):
    response = llm(prompt, max_tokens=512, temperature=0.7)
    return response["choices"][0]["text"]

gr.Interface(fn=chat, inputs="text", outputs="text", title="MegaBeam Mistral 512K - GGUF").launch()