from llama_cpp.server.app import create_app, Settings from fastapi.responses import HTMLResponse import os app = create_app( Settings( n_threads=2, # set to number of cpu cores model="model/gguf-model.bin", embedding=True, chat_format="mistrallite" ) ) # Read the content of index.html once and store it in memory with open("index.html", "r") as f: content = f.read() @app.get("/", response_class=HTMLResponse) async def read_items(): return content if __name__ == "__main__": import uvicorn uvicorn.run(app, host=os.environ["HOST"], port=int(os.environ["PORT"]) )