There are 5 open-source alternatives to LlamaIndex in this directory. The most popular is RAGFlow with 77.8k GitHub stars, followed by LightRAG(25.9k stars). All are open source, free to use, and many can be self-hosted.
| # | Tool | Stars | Language | License |
|---|---|---|---|---|
| 1 | RAGFlow RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that combines advanced RAG methods with Agent capabilities, delivering a superior, context-rich experience for Large Language Models (LLMs). It enables efficient information retrieval and contextual augmentation, optimizing generative AI pipelines. | 77.8k | Python | Apache License 2.0 |
| 2 | LightRAG LightRAG is a simple and fast Retrieval-Augmented Generation (RAG) framework, designed to enhance large language models with efficient retrieval mechanisms. Developed as part of EMNLP 2025, it focuses on streamlined performance and ease of integration for natural language processing tasks. | 25.9k | Python | MIT License |
| 3 | Mastra mastra is a powerful TypeScript AI agent framework designed for building intelligent assistants, retrieval-augmented generation (RAG) systems, and advanced observability features. It supports integration with any large language model (LLM), including GPT-4, Claude, Gemini, and Llama. | 22.9k | TypeScript | Other |
| 4 | OpenMemory OpenMemory allows AI agents to augment their intelligence with real, persistent long-term memory—distinct from vector search or retrieval-augmented generation (RAG), providing true storage and recall for AI systems. | 2.5k | TypeScript | Apache License 2.0 |
| 5 | Openai-apps-sdk-examples A collection of example applications demonstrating the usage and features of the OpenAI Apps SDK, providing practical reference implementations for developers integrating with the SDK. | 2.2k | TypeScript | MIT License |
RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that combines advanced RAG methods with Agent capabilities, delivering a superior, context-rich experience for Large Language Models (LLMs). It enables efficient information retrieval and contextual augmentation, optimizing generative AI pipelines.
LightRAG is a simple and fast Retrieval-Augmented Generation (RAG) framework, designed to enhance large language models with efficient retrieval mechanisms. Developed as part of EMNLP 2025, it focuses on streamlined performance and ease of integration for natural language processing tasks.
mastra is a powerful TypeScript AI agent framework designed for building intelligent assistants, retrieval-augmented generation (RAG) systems, and advanced observability features. It supports integration with any large language model (LLM), including GPT-4, Claude, Gemini, and Llama.
OpenMemory allows AI agents to augment their intelligence with real, persistent long-term memory—distinct from vector search or retrieval-augmented generation (RAG), providing true storage and recall for AI systems.
A collection of example applications demonstrating the usage and features of the OpenAI Apps SDK, providing practical reference implementations for developers integrating with the SDK.
The top three are RAGFlow(77.8k stars), LightRAG(25.9k stars), Mastra(22.9k stars). All are open source and free to use.
Yes. Every tool listed here is open source and free to use in your own projects. Many can be self-hosted on your own infrastructure, which means no subscription fees and full control over your data.
Many of the alternatives listed are self-hostable. Each tool's page lists hosting details, system requirements, and licensing terms.
Get notified about new tools and updates to existing ones.