Skip to main content

· One min read

The LangChain tutorial is shaping up nicely with a variety of key topics now added:

  • Models: Detailed guidance on utilising text completion models and chat models from OpenAI and Hugging Face.
  • Chains: Insights into how to link various components to act as a unified entity.
  • Agents: An exploration into autonomous entities using large language models as reasoning systems and tools for interacting with the environment.
  • Vector Stores: A deep dive into stores that can hold embeddings (vectors symbolising the meaning of text), perform semantic similarity searches among them, and use discovered text as context when querying large language models.

· One min read

Hey there! I'm so glad you've stumbled upon PromptChap, a blog where we'll venture into a few corners of the rapidly evolving world of AI. First, I'll be crafting a tutorial about LangChain, a handy framework used to create applications that utilise large language models.

Why LangChain? Well, it's all about streamlining the process of building applications with large language models, like those GPT models from OpenAI. LangChain comes packed with a variety of tools that enable you to load data from sources such as web pages or PDF documents, store this data as vectors in vector stores, and use it as context when interacting with the large language model. Additionally, it offers a way to create agents—these nifty components provide information on tools that the large language model can choose to use, and they run autonomously until they accomplish their goal.

Stay tuned, as over the next few weeks, I'll be publishing several tutorials on how to make the most of LangChain.