← Back to all projects

LangChain
The broadest set of model and tool integrations in one library
OfficialMIT
What is LangChain?
Its real value is coverage: whatever model, vector store or API you need, an adapter probably already exists, which shortens the distance from idea to prototype. The abstractions have a reputation for indirection, so many teams use it for the integrations and reach for something more explicit once the control flow gets complicated.
What can you do with LangChain?
- Swap model providers in place —
init_chat_modeltakes a "provider:model" string and returns a chat model behind one interface, so call sites keep using.invoke()when the provider changes. - Start from prebuilt agent patterns — Deep Agents, a higher-level package built on LangChain, ships planning, subagents and file-system use as built-in capabilities instead of patterns you assemble yourself.
- Move to explicit orchestration — When branching and state outgrow the chain-level API, LangGraph is the sibling low-level framework for writing controllable agent workflows directly.
- Build the same design in TypeScript — LangChain.js is the equivalent JS/TS library, distributed as the
langchainnpm package alongside the Python one.
Before you choose LangChain
- The interface is the same across providers; what sits behind it is not. One integration's docs promise a feature its code does not implement, so check what you depend on against the provider you will actually ship.Reported in#32197
Star history
15 Aug to 30 Oct · +4.5k
144k149k
Frequently asked questions
Is LangChain free for commercial use?
LangChain is released under the MIT licence — OSI-approved open source, which permits commercial use.
How can LangChain be deployed?
LangChain is available as Self-hosted / Runs locally.