src/LlamaIndex/draft.py

from llama_index.llms.huggingface_api import HuggingFaceInferenceAPI
from dotenv import load_dotenv
import os


load_dotenv()
llm = HuggingFaceInferenceAPI(
    model_name="Qwen/Qwen2.5-Coder-32B-Instruct",
    temperature=0.7,
    max_tokens=20,
    token=os.getenv("HUGGINGFACE_TOKEN")
)


if __name__ == "__main__":
    # Example usage
    response = llm.complete("Hello, how are you?")