This was a frequently requested feature, and we’re excited to finally release it! By utilizing streaming, you’ll be able to work with LLM responses token by token, reducing the perceived latency of your applications.

→ View original post on X — @langchain