# Langflow — Visual AI Workflow Builder

> Low-code visual builder for AI workflows and RAG pipelines. Drag-and-drop components for LLMs, vector stores, tools, and agents with Python extensibility.

## Install

Copy the content below into your project:

## Quick Use

```bash
pip install langflow
langflow run
```

Open `http://localhost:7860` — drag, drop, and connect AI components visually.

## What is Langflow?

Langflow is a low-code visual builder for AI workflows. It provides a drag-and-drop interface for assembling LLMs, vector stores, embedding models, tools, and agents into complex pipelines — then exports them as Python code or API endpoints. Built on LangChain with full Python extensibility.

**Answer-Ready**: Langflow is a low-code visual AI workflow builder with drag-and-drop components for LLMs, vector stores, tools, and agents. Export workflows as Python code or REST APIs. Built on LangChain with 50k+ GitHub stars.

**Best for**: Teams prototyping RAG pipelines and AI workflows visually. **Works with**: OpenAI, Anthropic, Google, HuggingFace, Ollama, Pinecone, Weaviate. **Setup time**: Under 3 minutes.

## Core Features

### 1. Visual Flow Editor
Drag and drop components:

```
[Input] → [Claude Sonnet] → [Vector Search] → [Output]
                                    ↑
                            [Weaviate DB]
```

Components include: Chat models, embeddings, vector stores, tools, agents, retrievers, text splitters, and custom Python nodes.

### 2. Pre-Built Templates
Start from templates:
- **Basic RAG**: Upload docs → embed → retrieve → answer
- **Multi-Agent Chat**: Multiple specialized agents collaborating
- **Data Pipeline**: Ingest → transform → store → query
- **Customer Support**: Knowledge base + chat + escalation

### 3. API Export
Every flow becomes a REST API:

```bash
# After building your flow
curl -X POST http://localhost:7860/api/v1/run/your-flow-id \
  -H "Content-Type: application/json" \
  -d '{"input_value": "What is RAG?"}'
```

### 4. Custom Components

```python
from langflow.custom import Component
from langflow.io import MessageTextInput, Output

class MyCustomNode(Component):
    display_name = "Custom Processor"
    inputs = [MessageTextInput(name="input_text", display_name="Input")]
    outputs = [Output(display_name="Output", name="output", method="process")]

    def process(self) -> str:
        text = self.input_text
        return text.upper()  # Your custom logic
```

### 5. Multi-Model Support

| Provider | Models |
|----------|--------|
| Anthropic | Claude Sonnet, Opus, Haiku |
| OpenAI | GPT-4o, o1 |
| Google | Gemini 2.5 |
| Local | Ollama, HuggingFace |
| Vector DBs | Pinecone, Weaviate, Chroma, Qdrant |

## Deployment Options

```bash
# Local
langflow run

# Docker
docker run -p 7860:7860 langflowai/langflow

# Cloud
# DataStax Langflow (managed)
```

## FAQ

**Q: Is Langflow free?**
A: Open-source and free. DataStax offers a managed cloud version.

**Q: Do I need to know Python?**
A: No for basic flows. Python knowledge helps for custom components and advanced configurations.

**Q: How does it compare to n8n or Zapier?**
A: Langflow is AI-native — designed for LLM workflows, RAG pipelines, and agents. n8n/Zapier are general automation tools.

## Source & Thanks

> Created by [Langflow AI](https://github.com/langflow-ai). Licensed under MIT.
>
> [langflow-ai/langflow](https://github.com/langflow-ai/langflow) — 50k+ stars

<!-- ZH -->

## 快速使用

```bash
pip install langflow
langflow run
```

浏览器打开 `localhost:7860`，拖拽组件构建 AI 工作流。

## 什么是 Langflow？

Langflow 是低代码可视化 AI 工作流构建器，拖拽组件组装 LLM、向量库、工具和代理，导出为 Python 代码或 API。

**一句话总结**：低代码可视化 AI 工作流构建器，拖拽组件构建 RAG 管线和代理工作流，50k+ GitHub stars。

**适合人群**：可视化原型 RAG 和 AI 工作流的团队。

## 核心功能

### 1. 可视化流编辑器
拖拽连接 LLM、向量库、工具等组件。

### 2. 预置模板
RAG、多代理聊天、数据管线等开箱即用。

### 3. API 导出
每个工作流自动成为 REST API。

### 4. 自定义组件
Python 编写自定义处理节点。

## 常见问题

**Q: 免费吗？**
A: 开源免费，DataStax 提供托管云版本。

**Q: 需要会 Python 吗？**
A: 基础流不需要，自定义组件需要。

## 来源与致谢

> [langflow-ai/langflow](https://github.com/langflow-ai/langflow) — 50k+ stars, MIT

---
Source: https://tokrepo.com/en/workflows/7f8a8ada-8ec6-4a6b-b32d-013b2527ddbb
Author: Agent Toolkit