LazyLLM — AI Agent 工作流中文教程
LazyLLM — AI Agent 工作流中文教程 是 AI Skill Hub 本期精选Agent工作流之一。已获得 3.8k 颗 GitHub Star，综合评分 8.9 分，整体质量较高。我们强烈推荐将其纳入你的 AI 工具库，帮助提升工作效率。
LazyLLM — AI Agent 工作流中文教程 工作流的设计遵循"最小配置，最大复用"原则：核心逻辑已经封装好，用户只需配置自己的 API Key 和业务参数即可快速上手。工作流内置错误处理和重试机制，在网络波动或 API 限速等情况下仍能稳定运行，适合作为生产环境的自动化基础设施。
在实际部署时，建议先在测试环境中运行 3-5 次，验证各个环节的输出结果符合预期，再部署到生产环境。AI Skill Hub 评分 8.9 分，是同类 Agent 工作流中的精选推荐。
LazyLLM — AI Agent 工作流中文教程 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排，将复杂的多步骤任务拆解为清晰的自动化流程，实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成，适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
LazyLLM — AI Agent 工作流中文教程 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排，将复杂的多步骤任务拆解为清晰的自动化流程，实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成，适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
- 可视化 Agent 工作流编排，无需编写复杂代码
- 支持多步骤自动化任务链，实现全流程无人值守
- 与外部 API、数据库和第三方服务无缝集成
- 内置错误处理与自动重试机制，保障稳定运行
- 提供可复用的自动化模板，快速在同类场景部署
- 自动化日常重复性工作，将精力集中于创造性任务
- 构建数据采集 → 处理 → 输出的完整自动化管线
- 实现跨平台、跨系统的数据流转和业务协同
# 方式一：pip 安装（推荐）
pip install lazyllm
# 方式二：虚拟环境安装（推荐生产环境）
python -m venv .venv
source .venv/bin/activate # Windows: .venv\Scripts\activate
pip install lazyllm
# 方式三：从源码安装（获取最新功能）
git clone https://github.com/LazyAGI/LazyLLM
cd LazyLLM
pip install -e .
# 验证安装
python -c "import lazyllm; print('安装成功')"
- 访问 GitHub 仓库获取工作流文件
- 在对应平台（Dify / Flowise / Make 等）中找到「导入工作流」功能
- 上传工作流文件
- 按照提示配置必要的环境变量和 API Key
- 运行测试确认流程正常后投入使用
# 命令行使用
lazyllm --help
# 基本用法
lazyllm input_file -o output_file
# Python 代码中调用
import lazyllm
# 示例
result = lazyllm.process("input")
print(result)
# lazyllm 配置文件示例（config.yml） app: name: "lazyllm" debug: false log_level: "INFO" # 运行时指定配置文件 lazyllm --config config.yml # 或通过环境变量配置 export LAZYLLM_API_KEY="your-key" export LAZYLLM_OUTPUT_DIR="./output"
简介
Features
Convenient AI Application Assembly Process: Even if you are not familiar with large models, you can still easily assemble AI applications with multiple agents using our built-in data flow and functional modules, just like Lego building.
One-Click Deployment of Complex Applications: We offer the capability to deploy all modules with a single click. Specifically, during the POC (Proof of Concept) phase, LazyLLM simplifies the deployment process of multi-agent applications through a lightweight gateway mechanism, solving the problem of sequentially starting each submodule service (such as LLM, Embedding, etc.) and configuring URLs, making the entire process smoother and more efficient. In the application release phase, LazyLLM provides the ability to package images with one click, making it easy to utilize Kubernetes' gateway, load balancing, and fault tolerance capabilities.
Cross-Platform Compatibility: Switch IaaS platforms with one click without modifying code, compatible with bare-metal servers, development machines, Slurm clusters, public clouds, etc. This allows developed applications to be seamlessly migrated to other IaaS platforms, greatly reducing the workload of code modification.<br>
Unified User Experience for Different Technical Choices: We provide a unified user experience for online models from different service providers and locally deployed models, allowing developers to freely switch and upgrade their models for experimentation. In addition, we also unify the user experience for mainstream inference frameworks, fine-tuning frameworks, relational databases, vector databases, and document databases.<br>
Efficient Model Fine-Tuning: Support fine-tuning models within applications to continuously improve application performance. Automatically select the best fine-tuning framework and model splitting strategy based on the fine-tuning scenario. This not only simplifies the maintenance of model iterations but also allows algorithm researchers to focus more on algorithm and data iteration, without handling tedious engineering tasks.<br>
Feature Modules
9.2.1 RAG - 9.2.1.1 Engineering - Integrate LazyRAG capabilities into LazyLLM (V0.7) - ✅ Extend RAG's macro Q&A capabilities to multiple knowledge bases (V0.6) - ✅ RAG modules fully support horizontal scaling, supporting multi-machine deployment of RAG algorithm collaboration (V0.6) - ✅ Integrate at least 1 open-source knowledge graph framework (V0.6) - Support common data splitting strategies, no less than 20 types, covering various document types (V0.6 - v0.8) - 9.2.1.2 Data Capabilities - Table parsing (V0.6 - 0.7) - CAD image parsing (V0.7 -) - pretrain data processing (V0.8) - 9.2.1.3 Algorithm Capabilities - Support processing of relatively structured texts like CSV (V0.7) - Multi-hop retrieval (links in documents, references, etc.) (V0.7) - Information conflict handling (V0.7) - AgenticRL & code-writing problem-solving capabilities (V0.7) - (new) AI Writter （V0.7 ） - (new) AI Review （V0.7 - 0.8 ）
9.2.2 Functional Modules - ✅ Support memory capabilities (V0.6) - Support for distributed Launcher (V0.7) - ✅ Database-based Globals support (V0.6) - ServerModule can be published as MCP service (v0.7) - Integration of online sandbox services (v0.7)
9.2.3 Model Training and Inference - ✅ Support OpenAI interface deployment and inference (V0.6) - Unify fine-tuning and inference prompts (V0.7) - Provide fine-tuning examples in Examples (V0.7) - Integrate 2-3 prompt repositories, allowing direct selection of prompts from prompt repositories (V0.7) - ✅ Support more intelligent model type judgment and inference framework selection, refactor and simplify auto-finetune framework selection logic (V0.6) - Full-chain GRPO support (V0.7)
9.2.4 Documentation - ✅ Complete API documentation, ensure every public interface has API documentation, with consistent documentation parameters and function parameters, and executable sample code (V0.6) - Complete CookBook documentation, increase cases to 50, with comparisons to LangChain/LlamaIndex (code volume, speed, extensibility) (V0.6 - v0.7) - ✅ Complete Environment documentation, supplement installation methods on win/linux/macos, supplement package splitting strategies (V0.6) - ✅ Complete Learn documentation, first teach how to use large models; then teach how to build agents; then teach how to use workflows; finally teach how to build RAG (V0.6)
9.2.5 Quality - ✅ Reduce CI time to within 10 minutes by mocking most modules (V0.6) - Add daily builds, put high-time-consuming/token tasks in daily builds (V0.7)
9.2.6 Development, Deployment and Release - Debug optimization (v0.7) - Process monitoring [output + performance] (v0.7 - 0.8) - Environment isolation and automatic environment setup for dependent training and inference frameworks (V0.7)
9.2.7 Ecosystem - ✅ Promote LazyCraft open source (V0.6) - Promote LazyRAG open source (V0.7) - ✅ Upload code to 2 code hosting websites other than Github and strive for community collaboration (V0.6)
LazyLLM: A Low-code Development Tool For Building Multi-agent LLMs Applications.
What can you build with Lazyllm
LazyLLM can be used to build common artificial intelligence applications. Here are some examples.
Installation
pip installation (recommended)
To install only lazyllm and necessary dependencies, you can use:
pip3 install lazyllmTo install lazyllm and all dependencies, you can use:
pip3 install lazyllm
lazyllm install fullInstallation from source
git clone git@github.com:LazyAGI/LazyLLM.git
cd LazyLLM
pip install -r requirements.txtInstallation on Windows or macOS
For installation on Windows or macOS, please refer to our tutorial
3.3 More Examples
For more examples, please refer to our official documentation Usage Examples Painting Master Multimodal Chatbot Knowledge Base Search Agent API Interaction Agent Tool-Calling Intelligent Agent
>>> lazyllm.demo.test()(1)
>>> lazyllm.demo.test_cmd(launcher=launchers.slurm)(2)
set environment variable: LAZYLLM_OPENAI_API_KEY=xx
or you can make a config file(~/.lazyllm/config.json) and add openai_api_key=xx
import lazyllm chat = lazyllm.OnlineChatModule() lazyllm.WebModule(chat).start().wait()
If you want to use a locally deployed model, please ensure you have installed at least one inference framework (lightllm or vllm), and then use the following code
Component
A Component is the smallest execution unit in LazyLLM; it can be either a function or a bash command. Components have three typical capabilities: 1. Cross-platform execution using a launcher, allowing seamless user experience: - EmptyLauncher: Runs locally, supporting development machines, bare metal, etc. - RemoteLauncher: Schedules execution on compute nodes, supporting Slurm, SenseCore, etc. 2. Implements a registration mechanism for grouping and quickly locating methods. Supports registration of functions and bash commands. Here is an example:
```python import lazyllm lazyllm.component_register.new_group('demo')
@lazyllm.component_register('demo') def test(input): return f'input is {input}'
@lazyllm.component_register.cmd('demo') def test_cmd(input): return f'echo input is {input}'
Module
Modules are the top-level components in LazyLLM, equipped with four key capabilities: training, deployment, inference, and evaluation. Each module can choose to implement some or all of these capabilities, and each capability can be composed of one or more components. As shown in the table below, we have built-in some basic modules for everyone to use.
| Function | Training/Fine-tuning | Deployment | Inference | Evaluation | |
|---|---|---|---|---|---|
| ActionModule | Can wrap functions, modules, flows, etc., into a Module | Supports training/fine-tuning of its Submodules through ActionModule | Supports deployment of its Submodules through ActionModule | ✅ | ✅ |
| UrlModule | Wraps any URL into a Module to access external services | ❌ | ❌ | ✅ | ✅ |
| ServerModule | Wraps any function, flow, or Module into an API service | ❌ | ✅ | ✅ | ✅ |
| TrainableModule | Trainable Module, all supported models are TrainableModules | ✅ | ✅ | ✅ | ✅ |
| WebModule | Launches a multi-round dialogue interface service | ❌ | ✅ | ❌ | ✅ |
| OnlineChatModule | Integrates online model fine-tuning and inference services | ✅ | ✅ | ✅ | ✅ |
| OnlineEmbeddingModule | Integrates online Embedding model inference services | ❌ | ✅ | ✅ | ✅ |
- 需要让 Claude / Cursor 操作本地工具的 AI 工程师
- 构建多智能体协作系统的 Agent 开发者
- 构建企业知识库 / RAG 检索应用的团队
- 做语音类 AI 产品的开发者
- 配置 MCP 服务器时建议使用 stdio 传输 + JSON-RPC，避免暴露公网
- 生产部署优先使用 Docker Compose 隔离依赖，并挂载 volume 持久化数据
- 分块大小建议 256-512 tokens，向量库优选 pgvector 或 Qdrant
- Agent 任务先做 dry-run 验证工具调用链，再开启自主执行
- API key 直接提交到 git 仓库（请用 .env 并加入 .gitignore）
- MCP 配置路径拼错或权限不足，重启 Claude Desktop 才生效
- 容器内无法访问宿主机 localhost — 使用 host.docker.internal
- embedding 模型与查询模型不一致导致检索失效
- Python 依赖冲突：建议用 venv / uv 隔离环境
- Docker：LazyLLM 提供官方镜像，docker compose up 一键启动
- 云端托管：可放在 Vercel / Railway / Fly.io 等 PaaS 平台
- 可视化 Agent 工作流编排，无需编写复杂代码
- 支持多步骤自动化任务链，实现全流程无人值守
- 与外部 API、数据库和第三方服务无缝集成
- 内置错误处理与自动重试机制，保障稳定运行
- 提供可复用的自动化模板，快速在同类场景部署
- 自动化日常重复性工作，将精力集中于创造性任务
- 构建数据采集 → 处理 → 输出的完整自动化管线
- 实现跨平台、跨系统的数据流转和业务协同
- +Apache-2.0 协议，可免费商用
- +AI Skill Hub 精选推荐
- +大幅减少重复性人工操作
- +可视化流程，清晰直观
- +可扩展性强，支持复杂场景
- −初始配置和调试需投入一定时间
- −强依赖外部服务的稳定性
- −复杂场景需具备一定技术基础
AI Skill Hub 为第三方内容聚合平台，本页面信息基于公开数据整理，不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后，再部署至生产环境，并做好必要的安全评估。
✅ Apache 2.0 — 宽松开源协议，可商用，需保留版权声明和 NOTICE 文件，含专利授权条款。
经综合评估，LazyLLM — AI Agent 工作流中文教程 在Agent工作流赛道中表现稳健，质量优秀。如果你已有明确的使用需求，可以直接上手体验；如果还在评估阶段，建议对比同类工具后再做决策。
| 原始名称 | LazyLLM |
| 原始描述 | Easiest and laziest way for building multi-agent LLMs applications. |
| Topics | agentsai-agentdatadeep-learningdocumentation-toolfinetuningllm-app |
| GitHub | https://github.com/LazyAGI/LazyLLM |
| License | Apache-2.0 |
| 语言 | Python |
收录时间：2026-05-22 · 更新时间：2026-05-22 · License：Apache-2.0 · AI Skill Hub 不对第三方内容的准确性作法律背书。