Prompt Engineer

Things have changed. This should work... Add the following environment variable to the pod environment variable called UI_ARGS to your pod with a value of --extensions openai --api-port 5001.

Yes. I shall look into FastAPI

Edit the memgpt_aget.py and define the generate_reply_for_user_message method to be async. I.e. "async def generate_reply_for_user_message....". This will fix this error.

Point well taken!

Try these end points http://127.0.0.1:5001/v1 http://127.0.0.1:5000/v1