Prasad Chalasani
Prasad Chalasani
@badal8381 did you get a chance to look at this?
add test in `test_vector_stores.py`
@Mohannadcse let's revisit this later. There may be a way to have it create an AzureGPT instance based on `type`.
Should implement an interface very similar to this: https://github.com/langroid/langroid/blob/main/langroid/vector_store/qdrantdb.py
> does one have to load all the vector databases, search engines etc etc on startup? Short answer is No. We have to put some time into deciding what we...
> Maybe there is a way to tag a tool such that it adds "LLM" as a recipient? Indeed, there's a SEND_TO routing you can add as prefix to the...
> atleast knocking off `mkdocs`, search engines and vectordb's would cut it to a third in size! Wow, surprising to hear that, thanks much for digging into this! We'll think...
Incidentally, we don't want to enforce that after `agent_response()` the result should always be handled by `llm_response()` -- one reason is that the result of a tool could be a...
There is another param in the `OpenAIGPTConfig`: `chat_context_length`, which defines the total (input + output) context length, and it defaults to 1024. When I ran this paper thru your settings,...
So... large number of dependencies and slow imports may be orthogonal issues. Reducing deps is good but won't speed up the imports, which is simply due to the large number...