LangChain
lastmod 2024-12-01created 2023-03-2214 edits
🦜🔗 Build context-aware reasoning applications
A library that connects LLMs to the outside world
Features
- Helps app developers use LLMs in more powerful ways across many fields
- LLMs and prompts: prompt management, prompt optimization, a common interface for LLMs
- Chains: chain multiple LLM calls together
- Data Augmented Generation: fetch and process information from external data sources. Can summarize external text or do QA
- Agents: an interface for agents that decide which action to take, take the action, observe, and repeat until done
- Memory: store state between chain/agent calls. Supports memory features through a standard interface
- Evaluation: evaluate with traditional metrics using language models
Ecosystem
- 🦜️🔗 LangChain - LangChain is a framework for developing applications powered by LLMs (large language models).
- 🦜🛠️ LangSmith - Helps you trace and evaluate language model applications and intelligent agents to move from prototype to production.
- 🦜🕸️ LangGraph - Build stateful, multi-actor applications with LLMs, built on top of (and intended to be used with) LangChain primitives.
- 🦜🏓 LangServe - Deploy LangChain runnables and chains as REST APIs.
Alternatives
- FlowiseAI
- Auto-GPT
- AgentGPT
- BabyAGI
- LangDock
- GradientJ
- TensorFlow
- LlamaIndex
Reported problems
See also
- LLM
- LangFlow - GUI for LangChain
- Ollama - Get up and running with Llama 2 and other large language models locally
- Retrieval-Augmented Generation (RAG; Korean: 검색 증강 생성)
- LocalAI
- LM Studio
Favorite site
- Github - teddylee777/langchain-kr
- A Korean tutorial written based on the official LangChain Document, Cookbook, and other practical examples. Through this tutorial you can learn how to use LangChain more easily and effectively.