Term
LangChain
Open-source framework that chains LLMs with data sources, tools and memory into applications — arguably the best-known pipeline library for AI apps.
LangChain — in more detail
LangChain is a framework (available for Python and JavaScript/TypeScript) that wires the building blocks of an LLM application — prompts, models, vector stores, tools, agents, memory — into reusable pipelines. At its core sits the LangChain Expression Language (LCEL), which composes components via a pipe operator into declarative chains. The ecosystem is rounded out by LangSmith (observability) and LangGraph (stateful agents as a graph).
Example / practical context
A typical LangChain application is a RAG system: documents are loaded, split into chunks, embedded and stored in a vector database such as Chroma or Pinecone. At runtime a retriever component pulls the relevant chunks, a prompt template combines them with the user’s question, and an LLM produces the answer. Tool-using agents — LLMs that can call functions like search, calculators or API endpoints — are another common use case.
Delineation from similar terms
LlamaIndex has a similar ambition but leans more heavily on data indexing and retrieval. Direct SDKs such as the openai or anthropic packages are thinner and have no pipeline abstraction — good for simple scripts, but without memory, routing or tool orchestration.
Entdecke mehr
Saving Tokens with Claude: 6 Principles That Make Experts Twice as Fast
How I turned my CLAUDE.md from a style guide into a token budget — 6 principles for lower cost, less waiting, and more honest reporting.
GlossarDSPy
DSPy is an open-source Python framework from the Stanford NLP group that lets you program LLM applications instead of hand-crafting prompts. Tasks are described through declarative signatures and modules; optimizers automatically derive effective prompts and examples.
LexikonPipeline Frameworks — LangChain, LlamaIndex and when you actually need them
LangChain, LlamaIndex, LangGraph and Haystack compared. What they're built for, when rolling your own pays off — and the criticisms worth taking seriously.