Retry Error with Urllib. I see the starting new connection and the model takes forever to reply

API
1

Converted retries value: 2 → Retry(total=2, connect=None, read=None, redirect=None, status=None)

Starting new HTTPS connection (1): api openai com:443

I keep getting this error and model takes forever to reply. Can I please get some help here?

2

Hi,

You seem to have forgotten to include any context, could you explain what this is in connection with? is this an OpenAI product you are making use of?

3

Yep I’m sorry.

So I’m using the GPT3.5 Turbo API and I use it with LangChain.

openai_llm = ChatOpenAI(model_name=‘gpt-3.5-turbo-1106’, streaming=True, callbacks=[StreamingStdOutCallbackHandler()],
temperature=0.5, max_retries=0)

I use this in my code and I see that sometime the model response is really fast and sometimes I just gives the Retry message in the console and doesn’t respond for 20 or 30 minutes.

I want to figure out if there is a way to avoid the Retry error and also improve model response

4

Well, in your example you have retries set to 0 so there will be no retries unless you have coded that in yourself, so maybe setting that to 3 as a test.