# LangGraph review

> LangGraph is a low-level orchestration framework and runtime for building, managing, and deploying long-running, stateful agents in Python or JavaScript.

Devtoolwatch Board, the week of Oct 25–31, 2026. Canonical page: https://index.devtoolwatch.com/tools/langgraph/

## TL;DR

LangGraph is a low-level orchestration framework for long-running, stateful agents. It is for developers who want fine-grained control and can mix deterministic, hand-coded steps with LLM-driven steps in one graph. LangGraph provides durable execution, human-in-the-loop, memory, and streaming. It works with or without LangChain.

## LangGraph at a glance

- **Category:** [Agent frameworks](https://index.devtoolwatch.com/categories/agent-frameworks/)
- **Maintained by:** LangChain (organization, [@langchain-ai](https://github.com/langchain-ai) on GitHub)
- **License:** MIT
- **Language:** Python, Makefile, TypeScript
- **Repository:** [github.com/langchain-ai/langgraph](https://github.com/langchain-ai/langgraph)
- **Website:** [docs.langchain.com/oss/python/langgraph](https://docs.langchain.com/oss/python/langgraph/)
- **npm package:** [@langchain/langgraph](https://www.npmjs.com/package/@langchain/langgraph) 1.5.2
- **PyPI package:** [langgraph](https://pypi.org/project/langgraph/) 1.2.16
- **Latest release:** [@langchain/langgraph 1.5.2](https://www.npmjs.com/package/@langchain/langgraph) on npm, Oct 21, 2026
- **First published:** Aug 9, 2023

## What LangGraph does

- LangGraph provides durable execution, so agents persist through failures and resume from where they left off.
- LangGraph supports human-in-the-loop oversight by letting you inspect and modify agent state at any point during execution.
- LangGraph offers short-term working memory and long-term persistent memory across sessions.
- LangGraph mixes deterministic, hand-coded steps with LLM-driven agentic steps in the same graph.
- LangGraph supports streaming, interrupts, time travel, subgraphs, and persistence through checkpointers and stores.
- LangGraph offers both a Graph API and a Functional API for defining workflows.
- LangGraph includes a Postgres checkpointer with a prune option for PostgresSaver and AsyncPostgresSaver.

### How it works

You install LangGraph as a Python library or use the JS/TS library LangGraph.js. You define agents as graphs, and persistence comes from checkpointers such as SQLite or Postgres. LangGraph runs standalone. It can also pair with LangChain for model and tool integrations, LangSmith for tracing and evaluation, and LangSmith Deployment, which the langgraph-cli deploy command targets.

## How to install LangGraph

**With pip**

```
pip install -U langgraph
```

### Requirements

- The Python package requires Python 3.10 or later.
- The JavaScript package @langchain/langgraph requires Node.js 18 or later.

### First steps

1. Learn the basic components used to build agents, starting with models and tools, before using LangGraph.
2. Install LangGraph with pip, then follow the LangGraph Quickstart in the documentation.
3. Take the free LangChain Academy introductory course to learn LangGraph basics.

## Who LangGraph is for

- Developers building long-running, stateful agents that must survive failures and resume.
- Teams that need custom workflows that combine predictable, auditable steps with model-driven steps.
- Applications that need human review or modification of agent state during execution.

### Limitations

- LangGraph is very low-level and focused only on orchestration. The docs recommend LangChain agents or Deep Agents for beginners.
- LangGraph's debugging visualization and managed deployment come from LangSmith, a separate product.

## Usage data (the week of Oct 25–31, 2026)

- Board rank: #79 overall, #5 in Agent frameworks, momentum score 68
- Downloads that week: 18.5M (+10% vs 4-week average)
- New GitHub stars per week (last 12 weeks): Aug 9: 520; Aug 16: 511; Aug 23: 441; Aug 30: 432; Sep 6: 453; Sep 13: 451; Sep 20: 374; Sep 27: 375; Oct 4: 377; Oct 11: 423; Oct 18: 444; Oct 25: 429

## LangGraph alternatives

| Rank | Tool | Stars | Downloads/wk |
|---|---|---|---|
| #1 | [Harness²](https://index.devtoolwatch.com/tools/google-research-harness2/) | 21.7k | – |
| #2 | [HyperFrames](https://index.devtoolwatch.com/tools/heygen-com-hyperframes/) | 76.6k | 1.5M |
| #3 | [AgentScope](https://index.devtoolwatch.com/tools/agentscope-ai-agentscope/) | 35k | 51.4k |
| #4 | [AG-UI](https://index.devtoolwatch.com/tools/ag-ui/) | 16.9k | 4.6M |
| #6 | [Bifrost](https://index.devtoolwatch.com/tools/maximhq-bifrost/) | 9.3k | – |
| #7 | [assistant-ui](https://index.devtoolwatch.com/tools/assistant-ui/) | 12.9k | 2.8M |

## LangGraph FAQ

### Is LangGraph free and open source?

Yes. LangGraph's source code is on GitHub (langchain-ai/langgraph) under the MIT license, so you can use it for free.

### How do I install LangGraph?

Run `pip install -U langgraph`. It needs Node.js >=18. The install section above has the other options from the project's README.

### Is LangGraph still maintained?

Yes. The langchain-ai/langgraph repository was last updated on Oct 21, 2026, and the latest release, @langchain/langgraph 1.5.2 on npm, came out on Oct 21, 2026.

### How popular is LangGraph?

LangGraph has 44.2k GitHub stars as of Nov 1, 2026. It gained 429 new stars in the week of Oct 25–31, 2026, against a 12-week average of 436 a week. Its main package had about 18.5M downloads that week, +10% against its own 4-week average. That made it #5 among AI agent frameworks on the Devtoolwatch Board.

### LangGraph vs LangChain: how do they compare?

LangGraph and LangChain are both tracked by Devtoolwatch as AI agent frameworks. LangGraph has 44.2k GitHub stars and LangChain has 148.8k. In the week of Oct 25–31, 2026, LangGraph's main package had 18.5M downloads and LangChain's had 41.1M. [LangChain review](https://index.devtoolwatch.com/tools/langchain/)

### LangGraph vs CrewAI: how do they compare?

LangGraph and CrewAI are both tracked by Devtoolwatch as AI agent frameworks. LangGraph has 44.2k GitHub stars and CrewAI has 60.4k. In the week of Oct 25–31, 2026, LangGraph's main package had 18.5M downloads and CrewAI's had 589.4k. CrewAI is an open-source Python framework for building multi-agent workflows, using role-based agent Crews and event-driven Flows. [CrewAI review](https://index.devtoolwatch.com/tools/crewai/)

### What are the best alternatives to LangGraph?

Other AI agent frameworks on the Devtoolwatch Board in the week of Oct 25–31, 2026, by momentum: Harness² (#1), HyperFrames (#2), AgentScope (#3), AG-UI (#4), Bifrost (#6). The alternatives table above has their numbers.

### Are LangGraph and LangChain the same?

No, LangGraph and LangChain are separate products. LangChain is the agent framework with abstractions and integrations for models and tools. LangGraph is the orchestration runtime for durable execution, streaming, human-in-the-loop, and persistence. Both are built by LangChain Inc, and LangGraph can be used without LangChain.

### Is LangGraph a framework or a harness?

LangGraph is a low-level orchestration framework and runtime, not a harness. The docs describe Deep Agents as the agent harness built on top of LangGraph. Deep Agents adds planning, subagents, filesystem tools, and context management.

### Is LangGraph used for agentic AI?

Yes, LangGraph is built for building, managing, and deploying long-running, stateful agents, including multi-actor applications with LLMs.

### How does LangGraph relate to LangSmith?

LangGraph builds the agents, and LangSmith handles tracing, evaluation, prompts, and deployment across frameworks. LangSmith Studio offers visual prototyping, and LangSmith Deployment hosts long-running, stateful workflows.

### Can LangGraph be used with JavaScript?

Yes, LangGraph.js is the equivalent JS/TS library. It is published on npm as @langchain/langgraph and requires Node.js 18 or later.

## Sources

1. [langchain-ai/langgraph on GitHub](https://github.com/langchain-ai/langgraph)
2. [LangGraph README](https://github.com/langchain-ai/langgraph/blob/main/README.md)
3. [@langchain/langgraph on npm](https://www.npmjs.com/package/@langchain/langgraph)
4. [langgraph on PyPI](https://pypi.org/project/langgraph/)
5. [LangGraph website](https://docs.langchain.com/oss/python/langgraph/)
6. [Release cli==0.4.37 (Oct 19, 2026)](https://github.com/langchain-ai/langgraph/releases/tag/cli%3D%3D0.4.37)
7. [Release 1.2.16 (Oct 16, 2026)](https://github.com/langchain-ai/langgraph/releases/tag/1.2.16)
8. [Release sdk==0.4.8 (Oct 16, 2026)](https://github.com/langchain-ai/langgraph/releases/tag/sdk%3D%3D0.4.8)
9. [Hacker News: Deep Agent Code Capabilities](https://news.ycombinator.com/item?id=48769008)
10. [Devtoolwatch Board, the week of Oct 25–31, 2026](https://index.devtoolwatch.com/)