Best of Langchain — September 2024

1. 1
   Article

   LangChain·2y

   Announcing LangChain v0.3

   LangChain v0.3 has been released, introducing significant changes for both Python and JavaScript users. In Python, all packages are upgraded to Pydantic 2, with Pydantic 1 and Python 3.8 support ending in 2024. In the JavaScript ecosystem, LangChain packages now use '@langchain/core' as a peer dependency, and callbacks are non-blocking by default. Multiple deprecated entry points and objects have been removed. The release also includes new integrations, revamped documentation, additional utilities for chat models, and a simplified tool definition. Guides are available for migration, and LangGraph remains the recommended way to build agents.

   34
2. 2
   Article

   DevBlogs·2y

   Build a chatbot on your own data in 1 hour with Azure SQL, Langchain and Chainlit

   Creating a custom chatbot using your own data has become straightforward with the help of LangChain, Chainlit, and Azure SQL. By leveraging Azure SQL's new vector support, you can store and query data embeddings. LangChain assists in defining prompts for the chatbot and retrieving relevant session data from the database. Finally, the integration with Chainlit ensures seamless Conversational AI implementation. This process takes less than an hour and can be further expanded based on your requirements.

   25
3. 3
   Article

   Towards AI·2y

   Building a Conversational Medical Form with GPT-4, Langchain, and SQLite

   Imagine replacing traditional static medical forms with a conversational assistant that uses GPT-4, guided by Langchain, to understand and store patient symptoms in a SQLite database. This post guides you through creating an interactive AI-driven medical form that alleviates patient stress by simulating a more personalized and caring interaction.

   20
4. 4
   Article

   neo4j·2y

   GraphRAG Field Guide: Navigating the World of Advanced RAG Patterns

   GraphRAG refers to a set of Retrieval Augmented Generation (RAG) patterns using a graph structure for improved data retrieval. The post introduces various GraphRAG patterns, categorized into basic, intermediate, and advanced levels, each requiring unique pre-processing and graph patterns. GraphRAG techniques often leverage relationships within data to enhance retrieval effectiveness, usually involving client-side data structures and Python code. The post also encourages contribution to an open-source GraphRAG Pattern Catalog and provides resources for further learning and implementation.

   16

   1
5. 5
   Article

   The New Stack·2y

   How To Build a RAG Agent With Nvidia NIM and LangChain

   Learn how to build an intelligent application that combines Retrieval-Augmented Generation (RAG) and tool calling using Nvidia NIM and LangChain. This system provides real-time flight status updates and detailed baggage information by integrating these technologies. The guide covers prerequisites, setting up the environment, initializing the language model, creating a flight status tool, binding tools to the language model, document loading and processing, setting up vector stores, implementing retrieval functions, and building the RAG chain.

   14