Is it possible to return a stream as a result of prompty.execute

[Gisela Torres (0GiS0)]

Hi folks 👋🏻

I'm trying to call my models with Prompty from python without using Semantic Kernel or Langchain and I can using this code:

import os
import prompty
# import invoker
import prompty.openai
from dotenv import load_dotenv

load_dotenv()

# execute the prompt
response = prompty.execute("example.prompty",
                           configuration={
                               "name": os.getenv('GITHUB_MODELS_MODEL'),
                               "type": "openai",
                               "base_url": os.getenv('GITHUB_MODELS_API_URL'),
                               "api_key": os.getenv('GITHUB_MODELS_API_KEY')
                           })

print(response)

But I would like to get a stream instead of the whole response at once. Is that possible today with Prompty? How?

Thank you so much!