You used langchain for a simple replacement of OpenAI API calls — of course it will increase complexity for no benefit.
The benefits of langchain are:
(1) unified abstraction across multiple different models and
(2) being able to plug this coherently into one architecture.
If you’re just calling some OpenAI endpoints, then why use it in the first place?
The benefits of langchain are: (1) unified abstraction across multiple different models and (2) being able to plug this coherently into one architecture.
If you’re just calling some OpenAI endpoints, then why use it in the first place?