This guide explains LLM evaluation metrics, including measures for retrieval augmented generation and agents, with DeepEval examples.
AutoRAG
AutoRAG helps an agent search for information across files on a computer.
About AutoRAG
AutoRAG Agent is a librarian agent that searches for information across files on a computer. It reads the files and answers questions in writing, with every claim linked to the passage it came from. The repository hosts AutoRAG 2.0, which is the current focus of development. The desktop app presents this engine inside a file manager, and the same engine is also available through a terminal UI and a command-line interface.
The original Python-based AutoRAG, a tool for finding an optimal RAG pipeline for a dataset, has moved to the legacy/ directory. It is in maintenance mode, and new feature work is focused on AutoRAG Agent. Models can be run through a hosted AutoRAG plan or through the user's own provider key, which involves an external service. Pricing is not stated.
Key features
- Answers questions about local files, with each claim linked to its source passage
- Hybrid keyword and meaning search with a reranker
- Searches Slack, Discord, Telegram, Mail, GitHub, and Obsidian alongside files
- Flags duplicate and similar files as Latest, Similar, or Identical
- Quick Look previews for PDF, Office, HWP, code, and media files
- Scheduled re-indexing that runs while the desktop app is closed
Best for: People who want to ask questions about files and connected messages or mail stored on their own computer.
Good for
- File contents are not sent to a remote embedder, because embedding runs on the user's machine
- Answers cite the passages they are based on
- API keys and sign-in tokens are stored in the system's encrypted store, not in a config file
Watch out for
- The original AutoRAG is in maintenance mode, and new features focus on AutoRAG Agent
- Hosted plans and external provider keys send requests to outside services
- WhatsApp access is read-only
AutoRAG at a glance
- Category
- AI & Machine Learning › AI Data & Analytics
- Type
- Open-source project
- Pricing
- Not confirmed. Check the site
- Open source
- Not stated
- Language
- English
- Last checked
- Oct 6, 2026 · Online when checked
- Listed
- Oct 5, 2026
AutoRAG alternatives
Tools and sites you could use instead, ranked by how closely they match.
Giskard provides open source evaluation and testing tools for LLM agents.
Open source
24 alternatives
Vearch provides distributed vector search capabilities for applications built around artificial intelligence.
Open source
24 alternatives
Open-RAG-Eval evaluates retrieval-augmented generation without requiring golden answers.
Open source
24 alternatives
MinerU converts complex PDF and Office documents into Markdown or JSON for AI workflows.
DB-GPT is an open-source agentic data assistant for AI and data products.
Open source
24 alternativesPopular in AI Data & Analytics
Well-known listings from the same topic.
Dotgetdot.aiForvioforvio.comFluidAI Medicalfluidai.mdApache Zeppelinzeppelin.apache.orgSola Securitysola.securityQuadraticquadratichq.com
AutoRAG FAQ
Are file contents sent to an outside service?
File contents are not sent to a remote embedder, because document understanding runs locally with the qwen3-embedding-0.6b model. If you use the hosted AutoRAG plan or your own provider key for answers, requests go to that external service.
Where is the original AutoRAG tool?
The original Python-based AutoRAG now lives in the legacy/ directory of the repository. It is in maintenance mode and can still be installed with pip install AutoRAG.
What interfaces are available?
The desktop app provides a file manager with an ask feature. The terminal UI provides interactive search sessions through autorag tui, and the CLI supports scripts, indexing, and health checks.
Is something wrong or out of date?
Tell us and an editor will check it.
Own this website?
Claim this listing
Claim the listing and suggest updates yourself. An editor checks every claim first.