# Retrieval-Augmented Generation

> Retrieval, embeddings, and crawling tools that ground LLM answers in your data.

Navigation: [Index](../../llms.txt) · [HTML](https://awesomeopensource.dev/topic/rag)

Retrieval-augmented generation connects a language model to your documents, databases, or the web before it answers. This topic covers crawlers, vector search libraries, and complete RAG servers.

Use this comparison to assemble a pipeline for ingestion, embedding, retrieval, and answer generation, or to adopt an end-to-end service with an API.

## Feature comparison

| Project | Document ingestion | Vector database support | Hybrid search | HTTP API | Source citations | Agentic retrieval |
|---|---|---|---|---|---|---|
| [open-webui/open-webui](../open-webui-open-webui.md) | Yes | Yes | No | Yes | Yes | No |
| [langchain-ai/langchain](../langchain-ai-langchain.md) | – | – | – | – | – | Yes |
| [infiniflow/ragflow](../infiniflow-ragflow.md) | Yes | Yes | Yes | Yes | Yes | Yes |
| [CherryHQ/cherry-studio](../cherryhq-cherry-studio.md) | Yes | – | – | No | – | Yes |
| [topoteretes/cognee](../topoteretes-cognee.md) | Yes | Yes | Yes | Yes | – | – |
| [unclecode/crawl4ai](../unclecode-crawl4ai.md) | Yes | No | Yes | Yes | Yes | No |
| [run-llama/llama_index](../run-llama-llama-index.md) | Yes | Yes | Yes | No | Yes | Yes |
| [yifanfeng97/Hyper-Extract](../yifanfeng97-hyper-extract.md) | Yes | – | – | No | Partial | No |
| [typesense/typesense](../typesense-typesense.md) | No | Yes | Yes | Yes | No | No |
| [dappros/ethora](../dappros-ethora.md) | – | – | – | – | – | – |
| [memvid/memvid](../memvid-memvid.md) | – | No | No | – | No | No |
| [kelindar/search](../kelindar-search.md) | Yes | No | No | No | Yes | No |
| [FrigadeHQ/trench](../frigadehq-trench.md) | No | No | No | No | Yes | No |
| [raksbisht/sitechat](../raksbisht-sitechat.md) | Yes | Yes | No | Yes | Yes | No |
| [SciPhi-AI/R2R](../sciphi-ai-r2r.md) | No | No | No | No | No | Yes |

## Features

- **Document ingestion**: Loads and parses PDFs, Office files or web pages.
- **Vector database support**: Integrates with a vector store such as Qdrant, pgvector or Chroma.
- **Hybrid search**: Combines keyword and semantic retrieval.
- **HTTP API**: Exposes retrieval or chat over a REST API.
- **Source citations**: Returns citations or references for answers.
- **Agentic retrieval**: Lets an agent plan and run multiple retrieval steps before answering.

## 15 projects

- [open-webui/open-webui](../open-webui-open-webui.md) · Activity score 5.0/5 (Thriving)
- [langchain-ai/langchain](../langchain-ai-langchain.md) · Activity score 5.0/5 (Thriving)
- [infiniflow/ragflow](../infiniflow-ragflow.md) · Activity score 5.0/5 (Thriving)
- [CherryHQ/cherry-studio](../cherryhq-cherry-studio.md) · Activity score 5.0/5 (Thriving)
- [topoteretes/cognee](../topoteretes-cognee.md) · Activity score 5.0/5 (Thriving)
- [unclecode/crawl4ai](../unclecode-crawl4ai.md) · Activity score 4.9/5 (Thriving)
- [run-llama/llama_index](../run-llama-llama-index.md) · Activity score 4.9/5 (Thriving)
- [yifanfeng97/Hyper-Extract](../yifanfeng97-hyper-extract.md) · Activity score 4.9/5 (Thriving)
- [typesense/typesense](../typesense-typesense.md) · Activity score 4.6/5 (Thriving)
- [dappros/ethora](../dappros-ethora.md) · Activity score 4.6/5 (Thriving)
- [memvid/memvid](../memvid-memvid.md) · Activity score 3.2/5 (Maintained)
- [kelindar/search](../kelindar-search.md) · Activity score 3.2/5 (Maintained)
- [FrigadeHQ/trench](../frigadehq-trench.md) · Activity score 1.8/5 (Slowing down)
- [raksbisht/sitechat](../raksbisht-sitechat.md) · Activity score 1.7/5 (Slowing down)
- [SciPhi-AI/R2R](../sciphi-ai-r2r.md) · Activity score 1.6/5 (Abandoned)
