“They handle all the integrations with all models (DeepSeek, Llama, gpt, claude)”
AI SDK & Agent FrameworkMaker says so +1 · source ↗
LangChain
Framework for building LLM applications by composing prompts, models, retrievers, tools, and agents.
66 makers' products we track use LangChain. For AI SDK & agent framework it's the most chosen of 7 options: 66 of the 115 makers' products we know the AI SDK & agent framework of (57%). 265 open-source projects also declare it in their code. What makers like most: One interface across many model providers, so switching models, prompts and chains is cheap. Their most common complaint: Frequent breaking API changes force rewrites, and coding agents trained on older versions produce messy code for it. On GitHub, 6 public pull requests moved projects to LangChain from tools we list, and 10 moved them away.
Profile checked
Works with AI agents:llms.txtHow makers use LangChain
Makers' products we track use LangChain for AI SDK & agent framework (66), most often alongside Next.js (36%), OpenAI API (35%), Vercel (35%).
What LangChain is used for
LangChain vs the alternatives makers chose
For AI SDK & agent framework, 66 of the 115 makers' products we know the choice of picked LangChain (57%), the most of 7 options.
| Option | Makers' products | Open source | |
|---|---|---|---|
| 66 57% | Yes | ||
| 24 21% | Yes | Compare → | |
| 16 14% | Yes | Compare → | |
| 5 4% | Yes | ||
| 3 3% | Yes | Compare → | |
| 1 1% | Yes |
For AI SDK & agent framework: how many of the 115 makers' products we know the AI SDK & agent framework of chose each option.
Often paired with
Share of the 66 makers' products on LangChain that show this choice.Markets that build with it most
Product Hunt markets where LangChain is in far more of the stacks we know than across every market (1%).
When to use LangChain, and when not
Pick LangChain if
- You need many prebuilt connectors for models, tools and data sources. AI SDK & Agent Framework →
Something else fits better if
- Your app is TypeScript and you want agents without adding a Python service: AI SDK or Mastra.
- You want a small typed agent loop in Python, close to the raw API: Pydantic AI or OpenAI Agents SDK.
- Your agent works on files and runs commands, like a coding or research agent: Claude Agent SDK.
- Your agent is a long-running workflow with branches, saved state and human approval steps: LangGraph.
- The agent mostly retrieves and reasons over your own documents: LlamaIndex.
LangChain is an open-source framework, in Python and TypeScript, for building LLM applications and agents. It gives you one interface over many model providers and a ready-made loop for agents that call tools, so you are not writing provider-specific glue code.
More about LangChain
The center of it is create_agent: you pass a model as a provider:model string, a list of tools (plain functions), and a system prompt, then add middleware for things like guardrails, retries or routing. Each provider integration ships as its own package (langchain-openai, langchain-anthropic and so on). Agents run on LangGraph, the lower-level orchestration runtime from the same team, which you can use directly when you need a graph that mixes fixed steps with model-driven ones. Deep Agents adds planning, subagents and a virtual filesystem on top.
From LangGraph you get durable execution, persistence, streaming and human-in-the-loop pauses. Agents can call tools on MCP servers through an adapter, and setting two environment variables sends traces to LangSmith, the company's hosted platform. The library runs wherever your code runs; LangSmith can also host agents as Agent Servers in US, EU or APAC regions, or you can run those servers yourself with Docker or Kubernetes.
The stack is several layered products (Deep Agents, LangChain, LangGraph, LangSmith), and part of the learning curve is picking the right layer. Self-hosting LangSmith itself is an Enterprise add-on, and self-run Agent Servers are not meant for serverless platforms that scale to zero.
How LangChain itself is built
18 tools, from its own code, website and Product Hunt page.
LangChain pros and cons
What makers like most: One interface across many model providers, so switching models, prompts and chains is cheap. Their most common complaint: Frequent breaking API changes force rewrites, and coding agents trained on older versions produce messy code for it.
Pros
- One interface across many model providers, so switching models, prompts and chains is cheap. PHPH 2PH 3
- A large ecosystem of document loaders, chunkers, retrievers and tool integrations for RAG and agents. PHPH 2PH 3
- LangSmith tracing and evaluation show what an agent did and where it went wrong. PHPH 2PH 3
Cons and watch-outs
- Frequent breaking API changes force rewrites, and coding agents trained on older versions produce messy code for it. HNHN 2HN 3
- Layers of abstraction make debugging slow, and many developers find calling model APIs or lighter libraries simpler. HNHN 2HN 3HN 4
- Its built-in patterns for subagents and deep research lag current practice, such as handing off context through a file system. HNHN 2
Themes that recur in makers' words and Hacker News comments, each linked to what it summarises, with how Product Hunt tags its reviews.
4.9★ from 116 reviews on Product Hunt ↗Reliability and open issues
The most wanted open issue on GitHub has 37 👍.
Most wanted on GitHub
- The batch method from ChatModels and all the Runnables does not really support the OpenAI batch API.👍 37 · opened Dec 2024 · active Aug 2026
- Support dynamic tool addition/removal after agent creation and in middleware👍 16 · opened Nov 2025 · active Sep 2026
- Prompts and Resources auto-discovery👍 16 · opened Apr 2025 · active Sep 2026
- [Feature Request] Native Support for MCP Code Execution (Programmatic Tool Calling) & Lazy Result Loading👍 15 · opened Nov 2025 · active Jul 2026
- Expose Deep Agent Harness as Composable LangGraph Primitives👍 13 · opened Mar 2026 · active Sep 2026
Who switches
More projects move away from LangChain: 6 pull requests in, 10 out; the most come from LangGraph and leave for LangGraph.
LangGraph → LangChain3+ PRs
- Slim down qvac-langgraph to a pure LangChain adapterSpaceUY/qvac-meridian-challenge · 2026-10-01
- docs(cli): rename langgraph template to langchain in package READMEsassistant-ui/assistant-ui · 2026-07-04
- feat(cli): rename the langgraph starter template to -t langchainassistant-ui/assistant-ui · 2026-06-19
AI SDK → LangChain3 PRs
- Migrate from Vercel AI SDK to LangChain for chat infrastructurekarthicksakkaravarti/projectX · 2026-04-28
- refactor: migrate from Vercel AI SDK to LangChain + LangGraphjjinendra3/AceFrame · 2025-08-17
- Moving Backend LLM Integration to from Vercel AI SDK to LangChainwesfloyd/vibe-avs-builder · 2025-04-29
LangChain → LangGraph6+ PRs
- feat(langchain): bridge MCP elicitation to LangGraph interruptslangchain-ai/langchain · 2026-08-20
- Migrate LLM calls to LangChain + orchestration to LangGraphKomali612/cicd-bootstrap · 2026-08-07
- archiving current langchain demo to start langgraph implementationyajji-k/ai-engineering-workspace · 2026-07-30
- Shifted chatbot form langchain to langgraph. Improved the workflow, M…Nouman64-cat/insurance-ai · 2026-07-23
- docs: add missing langchain to the LangGraph Python install commandComposioHQ/composio · 2026-07-15
- feat: migrate routing workflow from langchain to langgraphChikkalaSukrutiNaidu/multimodal-rag-bot · 2026-06-15
LangChain → AI SDK4 PRs
- server/llm: migrate from LangChain to Vercel AI SDKsagemathinc/cocalc · 2026-04-14
- Migrate LLM evaluation from LangChain to Vercel AI SDKsagemathinc/cocalc · 2026-04-10
- Migrate LangChain to Vercel AI SDKriccjohn/grimoire-oracle-web · 2026-03-02
- Migrate from LangChain to Vercel AI SDKliam-hq/liam · 2025-05-08
Public pull requests on GitHub since Oct 2024 whose title says "X to Y" — real code changes moving a project from one tool to another, by developers in general.
Who uses it
66 makers' products, each linked to the source that shows it, and 265 open-source projects that declare it in their code. The best known on Product Hunt: STORI (2,910 followers), Databox (2,312 followers), Permit.io (1,970 followers), reflex (1,848 followers), Prefactor (1,415 followers). 1 of the makers' products is gone now (domain lapsed, up for sale or taken down); it chose LangChain while live.
“Core framework for LLM orchestration, tool binding, prompt management, and embeddings. Powers the entire AI layer across all services.”
AI SDK & Agent FrameworkMaker says so · source ↗“The whole ecosystem has been a great help till now. Even though we often have to develop our own. Langsmith is one of our go to tool for tracing and evaluating.”
AI SDK & Agent FrameworkMaker says so · source ↗“Quick, scalable, and simple to implement.”
AI SDK & Agent FrameworkMaker says so · source ↗“The @tiptap-pro/ai-toolkit-langchain package provides tool definitions you can add to your AI agent built with LangChain.js.”
AI SDK & Agent FrameworkMaker says so · source ↗“LangGraph by LangChain made it possible to design the agentic workflow as state-graphs providing greater control and reliability. Also, LangSmith for observabilty of the Agentic actions.”
AI SDK & Agent FrameworkMaker says so · source ↗“LangSmith has been a lifesaver for testing and debugging our prompt flows. It gave us the visibility we needed to make Reflex AI Builder feel smooth and reliable. Super helpful when working with complex LLM chains.”
AI SDK & Agent FrameworkMaker says so · source ↗“The best framework for inferencing LLM models.”
AI SDK & Agent FrameworkMaker says so · source ↗“We evaluated several orchestration frameworks before choosing LangGraph for Artifacts. Generating a document isn't one call and done, it's a multi-step flow: pull the data, structure the layout, apply styling, and hold state across follow-ups like "turn this into slides." The alternatives either abstracted too much away or couldn't handle that stateful, multi-step process cleanly. LangGraph gave us explicit control over each step, so a report doesn't drift into something the user didn't ask for halfway through generation.”
AI SDK & Agent FrameworkMaker says so · source ↗“At the start, we combine LangChain functionality, and then we improve our process so that we can now choose whether to use it or not”
AI SDK & Agent FrameworkMaker says so · source ↗“We use Langchain to orchestrate our entire agent framework — from the AI inbox and SDR copilots to outbound flows and long-term memory. It helps manage complex tool use, contextual reasoning, and multi-turn interactions across our platform. Langchain has been critical to scaling structured LLM workflows with modular, maintainable logic. Grateful to the Langchain team and open-source contributors (part of pour team) pushing this ecosystem forward.”
AI SDK & Agent FrameworkMaker says so · source ↗“Nifty way to fine-tune LLMs to suit our AI Agent use-case.”
AI SDK & Agent FrameworkMaker says so · source ↗“Connects all the moving parts of our chat stack.”
AI SDK & Agent FrameworkMaker says so · source ↗“The atoms to our molecules.”
AI SDK & Agent FrameworkMaker says so · source ↗“Langgraph for managing our AI flows”
AI SDK & Agent FrameworkMaker says so · source ↗“Langchain helps us define parts of our agent.”
AI SDK & Agent FrameworkMaker says so · source ↗“Awesome tooling to build agents”
AI SDK & Agent FrameworkMaker says so · source ↗“LangGraph's graph-based architecture with nodes and edges was the key on how we coordinate complex AI workflows. We're orchestrating multi-agent conversations where each node handles specific tasks (analysis, decision-making, action execution) and edges manage the flow between them. Memory lets us build operations that maintain context across conversations, while parallel execution nodes handle multiple AI operations simultaneously. The Pydantic formatted outputs ensure type-safe data flow between nodes, and the checkpoint system with PostgreSQL persistence lets us maintain state across these graph executions, so conversations can resume seamlessly even after server restarts.”
AI SDK & Agent FrameworkMaker says so · source ↗“Langchain was again crucial in building our browser agent which has improved a ton in terms of reliability”
AI SDK & Agent FrameworkMaker says so · source ↗“Using Langchain for Minicule Research AI Agents.”
AI SDK & Agent FrameworkMaker says so · source ↗“We use LangChain to build our RAG pipeline.”
AI SDK & Agent FrameworkMaker says so · source ↗“Made connecting our AI to real data feel intuitive, not intimidating.”
AI SDK & Agent FrameworkMaker says so · source ↗“Great for rapid and reliable AI development”
AI SDK & Agent FrameworkMaker says so · source ↗“Langgraph Studio helps us easily build and ideate agents”
AI SDK & Agent FrameworkMaker says so · source ↗“LangChain retrievers create endless opportunities to secure RAG”
AI SDK & Agent FrameworkMaker says so · source ↗“Our customers build on LangChain, so we do too, and it means we hit the same edges they hit before they have to tell us about them.”
AI SDK & Agent FrameworkMaker says so · source ↗“powers our AI workflows and makes them easy to adapt as we grow”
AI SDK & Agent FrameworkMaker says so · source ↗“Helped us deploy and test agents”
AI SDK & Agent FrameworkMaker says so · source ↗“Gives us flexibility to work with various model providers, avoiding vendor lockin.”
AI SDK & Agent FrameworkMaker says so · source ↗“Big thanks to LangChain for powering our AI development workflow! ⚡️ From chaining LLM calls to managing memory and integrations, LangChain makes building intelligent apps faster and more reliable. Couldn’t create without it! 🚀”
AI SDK & Agent FrameworkMaker says so · source ↗“The orchestration layer that ties Fonda's AI together. LangChain made it possible to build structured, multi-step workflows that remember context across the entire 14-step founder journey -- not just one session. Hard to build a co-founder with memory without it.”
AI SDK & Agent FrameworkMaker says so · source ↗“LangChain gives the right level of abstraction without hiding what’s going on. It lets us move fast, experiment, and still stay in control of how prompts and logic actually work, and it enables you to use a lot of different models inside TypeBoost. A perfect fit for a product that’s all about precision and customization.”
AI SDK & Agent FrameworkMaker says so · source ↗“Sara’s multi-agent architecture uses LangChain to orchestrate internal reasoning and task routing.”
AI SDK & Agent FrameworkMaker says so · source ↗“LangGraph (by LangChain) powers our agent orchestration layer. We evaluated building a custom state machine, but LangGraph's approach to cyclic graphs with built-in tool calling and checkpointing saved us months. The ability to define complex agent workflows with conditions, branches, and human-in-the-loop — while staying in TypeScript — was the deciding factor. It's the backbone behind our visual workflow builder.”
AI SDK & Agent FrameworkMaker says so · source ↗“Huge thanks to LangChain for helping us build Indigo’s seamless AI integration and powerful command features! 🚀”
AI SDK & Agent FrameworkMaker says so · source ↗“I used the lang graph extension to be able to design the complex inventor agent.”
AI SDK & Agent FrameworkMaker says so · source ↗“Powers the AI workflows to analyze web content, detect phishing, and contextual threats intelligently.”
AI SDK & Agent FrameworkMaker says so · source ↗“Great for building agents!”
AI SDK & Agent FrameworkMaker says so · source ↗“Leverages LangGraph for the AI data analyst agent.”
AI SDK & Agent FrameworkMaker says so · source ↗“Fantastic ability to trace and troubleshoot for AI-first applications”
AI SDK & Agent FrameworkMaker says so · source ↗“The ultimate enabler for in-browser, private RAG. This tool is the secret sauce that allowed me to generate high-quality vector embeddings directly in the browser via WebAssembly. It completely eliminated the need to spin up expensive cloud servers or compromise user privacy. If you are building local-first edge AI or hybrid search engines, this is the absolute gold standard you need in your tech stack.”
AI SDK & Agent FrameworkMaker says so · source ↗“Their modular tooling helped us prototype and test personality injection workflows early on.”
AI SDK & Agent FrameworkMaker says so · source ↗“LangSmith-style tracing so you can see what the agent did, why it did it, and where things went wrong.”
AI SDK & Agent FrameworkMaker says so · source ↗“Archie builds all interactions with LLM using Langchain. We evaluated multiple frameworks and even worked directly with vendors' API, but quickly realized that investing in Langchain would allow us to leverage a robust ecosystem. These days, it is pretty simple for our AI labs to switch prompts, sequences, LLM, and agents, given the framework's flexibility.”
AI SDK & Agent FrameworkMaker says so · source ↗“Langchain helps us define parts of our agent.”
AI SDK & Agent FrameworkMaker says so · source ↗“LangChain helped us orchestrate complex reasoning, memory, and planning steps behind our AI employee. It’s the brain behind turning a conversation into a functional website.”
AI SDK & Agent FrameworkMaker says so · source ↗“Langraph, helped us build our agent OFFY.”
AI SDK & Agent FrameworkMaker says so · source ↗None of these in makers' own words.
Open source: a project that declares LangChain as a dependency in its public code — verifiable, but not necessarily a live product.
What its projects tell coding agents
90 of the 154 open-source products built with LangChain that were worked on in the last 180 days keep instruction files for coding agents — 58%, against 53% of all the projects we read. AGENTS.md is in 73, CLAUDE.md in 57.
Alternatives to LangChain
All alternatives by situation →Further reading
Questions makers ask about LangChain
Is LangChain good for production?
66 makers' products we track run on it, and 1 of them have since gone (the product, not necessarily because of LangChain). On GitHub, 6 public pull requests moved projects to LangChain from tools we list, and 10 moved them away. The watch-out makers raise most: Frequent breaking API changes force rewrites, and coding agents trained on older versions produce messy code for it.
What is LangChain used for?
Framework for building LLM applications by composing prompts, models, retrievers, tools, and agents. Makers' products we track use it for AI SDK & agent framework (66).
What are the pros and cons of LangChain?
Pros makers name: One interface across many model providers, so switching models, prompts and chains is cheap. A large ecosystem of document loaders, chunkers, retrievers and tool integrations for RAG and agents. LangSmith tracing and evaluation show what an agent did and where it went wrong. Cons: Frequent breaking API changes force rewrites, and coding agents trained on older versions produce messy code for it. Layers of abstraction make debugging slow, and many developers find calling model APIs or lighter libraries simpler. Its built-in patterns for subagents and deep research lag current practice, such as handing off context through a file system.
What are the alternatives to LangChain?
For AI SDK & agent framework, the other options makers' products we track chose most are Mastra (24), LlamaIndex (16), AI SDK (5), CrewAI (3), LangGraph (1). Open source or self-hostable among them: Mastra, LlamaIndex, AI SDK, CrewAI, LangGraph. Each alternative, with the situation it fits, is on the LangChain alternatives page.
Do developers switch to or away from LangChain?
On GitHub, 6 public pull requests moved projects to LangChain from tools we list, most from LangGraph (3), AI SDK (3); 10 moved them away, most to LangGraph (6), AI SDK (4).
When should you use LangChain?
You need many prebuilt connectors for models, tools and data sources (AI SDK & agent framework). Something else fits better if: your app is TypeScript and you want agents without adding a Python service, pick AI SDK or Mastra; you want a small typed agent loop in Python, close to the raw API, pick Pydantic AI or OpenAI Agents SDK.
Does it work with TypeScript?
Yes. LangChain.js installs from npm as langchain plus @langchain/core and needs Node.js 22 or later, or Bun 1.0 or later. source ↗
Do I need a LangSmith account to use it?
No. Tracing to LangSmith is turned on by setting LANGSMITH_TRACING and an API key; agents run the same without them. source ↗
Where does LangSmith store my traces?
In the region you sign up in: the US (GCP Iowa or AWS Ohio), the EU (GCP Netherlands) or APAC (GCP Sydney). Only billing data is kept in the US regardless of region. source ↗
Can I self-host LangSmith?
Only as an add-on to the Enterprise plan. It installs on Kubernetes and needs ClickHouse, PostgreSQL and Redis, with blob storage recommended for production. source ↗
Can I deploy agents on my own infrastructure?
Yes. Package the agent as a Docker image and run it as a standalone Agent Server on Kubernetes, Docker or a VM, optionally still sending traces to LangSmith. The docs warn against serverless platforms, where scale-to-zero can lose tasks. source ↗
Will upgrades break my code?
Releases follow semantic versioning: stable APIs keep backward compatibility and breaking changes come only in major versions, with migration guides. APIs marked beta or alpha can still change. source ↗
Can agents use tools from MCP servers?
Yes. MCPAdapter in langchain[mcp] (version 1.4 or later) discovers an MCP server's tools and turns them into LangChain tools. The namespace is still in beta. source ↗
Is LangChain free?
Yes — there is a free tier a small product can run on; paid use starts at $39/seat/mo (LangSmith). source ↗
Is LangChain open source or self-hostable?
Open source, and you can self-host it. source ↗
Can AI coding agents work with LangChain?
It serves an llms.txt docs index.
Who uses LangChain?
66 makers' products we track, each with a source, and 265 open-source projects declare it in their code. The best known of them on Product Hunt: STORI (2,910 followers), Databox (2,312 followers), Permit.io (1,970 followers), reflex (1,848 followers), Prefactor (1,415 followers). source ↗