Once the response is received from the

The prompt templates by default specify that only answer based on the given context information. GPT-4 especially does a good job with following that

But if I want the answer from public domain if the answer is not in the trained documents what should be done ?

You can also specify that in the prompt templates or then you can use a separate query engine for those questions

Can you please redirect to any examples for this use case if you know any ?

You can use these: https://docs.llamaindex.ai/en/stable/module_guides/models/prompts.html****

Or build separate query engines: https://docs.llamaindex.ai/en/stable/examples/query_engine/RouterQueryEngine.html

I think this is a ready made implementation for those sorts of use-cases but it's for the chat engine: https://docs.llamaindex.ai/en/stable/examples/chat_engine/chat_engine_react.html

👍 thank u