Lemon AI: Full-stack Open-source Self-Evolving General AI Agent
This repository profile is provided by osrepos.com, an open source repository discovery platform.

Summary
Lemon AI is a pioneering full-stack, open-source, self-evolving general AI agent designed as a fully local alternative to cloud-based agentic platforms. It offers comprehensive capabilities for deep research, coding, and data analysis, ensuring complete privacy and zero cloud dependency by running entirely on local hardware. This agent features an integrated Code Interpreter VM sandbox for safe execution and an innovative AI Agent HTML Editor.
Repository Information
Topics
Click on any tag to explore related repositories
Use at your own risk
OSRepos shares public repositories for knowledge and discovery only. Any installation, execution, configuration, or use of code from these repositories is the user's own responsibility. Always review the repository, source code, dependencies, licenses, and security implications before running or installing anything. OSRepos is not responsible for issues, damages, or losses resulting from third-party repositories.
Introduction
Lemon AI is the first Full-stack Open-source Self-Evolving General AI Agent, offering a fully local alternative to Agentic platforms like Manus and Genspark AI. It features an integrated Code Interpreter VM sandbox for safe execution, empowering deep research, web browsing, viable coding, and data analysis entirely on your local hardware.
This agent supports planning, action, reflection, and memory functionalities using local LLMs (such as DeepSeek, Kimi, Qwen, Llama, Gemma, GPTOSS) via Ollama and VLLM, ensuring complete privacy and zero cloud dependency. For enhanced security, Lemon AI operates within a local Virtual Machine (VM) sandbox by Docker, protecting your machine's files and operating system. Additionally, it provides the flexibility to configure enhanced results using APIs from leading cloud models like Claude, GPT, Gemini, and Grok.
Installation
The easiest way to run Lemon AI is in Docker. Ensure you have Docker Desktop installed and configured for your operating system (MacOS, Linux, or Windows with WSL). A system with a modern processor and a minimum of 4GB RAM is recommended.
Here are the steps to get started:
docker pull hexdolemonai/lemon-runtime-sandbox:latest
docker run -it --rm --pull=always \
--name lemon-app \
--env DOCKER_HOST_ADDR=host.docker.internal \
--env ACTUAL_HOST_WORKSPACE_PATH=${WORKSPACE_BASE:-$PWD/workspace} \
--publish 5005:5005 \
--add-host host.docker.internal:host-gateway \
--volume /var/run/docker.sock:/var/run/docker.sock \
--volume ~/.cache:/.cache \
--volume ${WORKSPACE_BASE:-$PWD/workspace}:/workspace \
--volume ${WORKSPACE_BASE:-$PWD/data}:/app/data \
--interactive \
--tty \
hexdolemonai/lemon:latest make run
For more detailed instructions and prerequisites, refer to the official documentation.
Examples
Lemon AI supports unlimited task scenarios, making it highly versatile for various applications:
- Deep search and research reports: Conduct thorough investigations and generate comprehensive reports.
- Code generation and data analysis: Automate coding tasks and perform complex data analysis.
- Content creation and document processing: Generate diverse content and streamline document workflows.
A key feature is its Self-Evolving capability, where it creates a personal memory for every conversation, making the agent smarter and more effective with continuous use.
Furthermore, Lemon AI introduces the AI Agent HTML Editor, the world's first General AI Agent Editor. This editor empowers users to continuously edit, modify, and refine generated HTML pages. You can let AI modify any section content, insert new paragraphs, reformat the entire page, or use advanced edit mode for direct manual text adjustments, ensuring the results meet your exact satisfaction.
Why Use Lemon AI
- Complete Privacy and Local Execution: Unlike many agentic platforms, Lemon AI runs entirely on your local hardware, offering complete privacy and zero cloud dependency for sensitive tasks.
- Enhanced Security: It operates within a local Virtual Machine (VM) sandbox by Docker, which safely handles all code writing, execution, and editing tasks, protecting your machine's files and operating system.
- Cost-Effectiveness: Lemon AI dramatically reduces operational costs, with task execution costing significantly less than other agent products. It leverages open-source models like DeepSeekV3, making advanced AI capabilities accessible.
- Powerful and Flexible: This feature-rich framework includes virtual machine integration, code generation and execution, browser operations, web search, and multi-tool integration. Its highly adaptable architecture allows for custom modifications and extensions.
- Rapid Deployment: With one-click deployment options, Lemon AI ensures quick setup and immediate usage with minimal technical requirements, supporting various environments from personal computers to enterprise servers.
Links
- Get to know Lemon AI quickly: https://lemon-11.gitbook.io/lemonai
- Docker Quick Deployment: https://lemon-11.gitbook.io/lemonai/development-deployment-guidelines/docker-quick-deployment
- Documentation: https://document.lemonai.cc/
- Official Website: https://lemonai.cc/
- GitHub Repository: https://github.com/hexdocom/lemonai
- GitHub Discussions: https://github.com/hexdocom/lemonai/discussions
- X (Twitter): https://x.com/LemonAI_cc
- Discord: https://discord.com/invite/gjEXg4UBR4
Related repositories
Similar repositories that may be relevant next.

Feynman: The Open Source AI Research Agent
June 2, 2026
Feynman is an open-source AI research agent designed to automate and streamline complex research tasks. Built with TypeScript, it leverages multiple agents and tools to conduct in-depth investigations, literature reviews, and even experiment replications, providing source-grounded outputs.

CodeGraph: Supercharge AI Coding Agents with Semantic Code Intelligence
May 25, 2026
CodeGraph is a powerful, pre-indexed code knowledge graph designed to enhance AI coding agents like Claude Code, Cursor, and Codex. It significantly reduces token usage and tool calls, offering a faster and more cost-effective way for agents to understand codebases. This 100% local solution provides semantic code intelligence, improving agent efficiency and accuracy.

Agentic Inbox: A Self-Hosted Email Client with AI on Cloudflare Workers
May 18, 2026
Agentic Inbox is an innovative self-hosted email client that integrates an AI agent, running entirely on Cloudflare Workers. It provides a modern web interface for managing emails, enhanced by AI capabilities for reading, searching, and drafting replies. This project leverages Cloudflare's robust ecosystem, including Email Routing, Durable Objects, R2, and Workers AI, to deliver a powerful and secure email solution.

UI-TARS-desktop: The Open-Source Multimodal AI Agent Stack
May 6, 2026
UI-TARS-desktop is an open-source multimodal AI Agent stack from ByteDance, designed to connect cutting-edge AI models with agent infrastructure. It provides both Agent TARS, a general multimodal AI agent with CLI and Web UI, and UI-TARS Desktop, a native GUI agent for local and remote computer/browser control. This powerful tool aims to enable human-like task completion through rich multimodal capabilities and seamless integration with real-world tools.
Source repository
Open the original repository on GitHub.