Invoking langchain Chains and agents using REST API

[sasidharreddy25]

Checked

• I searched existing ideas and did not find a similar one
• I added a very descriptive title
• I've clearly described the feature request and motivation for it

Feature request

from langchain.chains import create_sql_query_chain
from langchain_openai import ChatOpenAI

llm = ChatOpenAI(model="gpt-3.5-turbo", temperature=0)
chain = create_sql_query_chain(llm, db)
response = chain.invoke({"question": "How many employees are there"})
response

The above code takes openai endpoint, key in ChatOpenAI object to call openai when chain is invoked

In our scenario due to security reasons we have API Gateway infront of openai, so we need to hit API Gateway using REST API call, then it will route our request to openai.

So we are invoking API Gateway using python:
response=request.post(endpoint(api gateway), data=data)

But how can we integrate this REST API code with current above langchain code, it only takes direct openai endpoint and keys, but that is not the scenario in our case

Motivation

Due to the change in the way we call openai this problem a raises, please show us some solution for this problem

Proposal (If applicable)

No response