Carrel
Carrel is a self-hosted knowledge base that lets AI agents search your own documents and their related facts for accurate answers.
What it is good for
- Giving an AI agent answers from your own product documents and files
- Searching contracts, research, and meeting notes across many files
- Finding source passages and images inside large document collections
- Keeping a searchable knowledge base updated as files change
At a glance
- Who it is for
- Setup
- Guided command-line setup
Level 4 of 5. You copy and paste commands into a terminal by following the instructions. Comfortable territory for vibe coders.
- Cost
- Free and open source
- Type
- Agent app
- Tasks
- Works with
- Claude Code, Codex, Neo4j, Qdrant, OpenSearch, MinerU
Momentum
- Rank this week
- Not ranked (last #66, week 42, 2026)
- Best rank
- #66 week 42, 2026
- GitHub stars
- 681
GitHub numbers as of Nov 2, 2026. Updated weekly.
Ranking history
Official links
Setup steps are on the official pages. We do not repeat them here, since they change.
Questions
- Is Carrel free?
- Yes. It is free and open source (MIT license).
- Do I need to know how to code to use Carrel?
- You do not need to write code, but setup means copying and pasting commands into a terminal by following its instructions.
- What does Carrel work with?
- According to its own documentation, it works with Claude Code, Codex, Neo4j, Qdrant, OpenSearch, and MinerU.
- What is Carrel good for?
- Giving an AI agent answers from your own product documents and files, searching contracts, research, and meeting notes across many files, finding source passages and images inside large document collections, and keeping a searchable knowledge base updated as files change.
Similar agents to Carrel
- MOVO
MOVO is a self-hosted platform that turns DeepSeek's agent runtime into a team workspace with knowledge search, research, and admin controls for organizations.
- GPT Researcher
GPT Researcher is an agent that runs deep web and local document research and writes cited reports for people who need factual summaries.
- AutoResearchClaw
AutoResearchClaw is an agent app that turns a research idea into a full academic paper with citations and experiments, for researchers and students.
- Local Deep Research
Local Deep Research is an agent app that runs multi-step research across the web, academic papers, and your own documents for people who want privacy and control.
- Claude Scholar
Claude Scholar is a set of skills and agents that helps computer science and AI researchers manage literature, experiments, and paper writing.
- Open Science Desktop
Open Science Desktop is a desktop app that runs AI research workflows from literature survey to paper, with a local, auditable record for researchers.