ibmtjbot/tjbot

Add new recipe for `conversation-watsonx`

jweisz opened this issue · 1 comments

jweisz commented

We should implement a new recipe to demonstrate a conversational LLM experience with TJBot. Now that there are chat-tuned models available (e.g. llama-2-70b-chat and granite-13b-chat-v1), this should work well.

Also, use langchain to drive the conversation! 😄

The code for this is working - I'm just having trouble testing the hardware so I'll continue to work on that