RAGFlow
A retrieval and agent platform for turning documents into usable model context. Its current documentation combines document processing, knowledge compilation, and agent workflows.
Utility: Knowledge and memory; kind: platform.
Stars: 93k+; Starred in September: 1k+.
Metrics updated: 31 Oct 2026. Profile reviewed: 31 Oct 2026.
Utilities
- Parse and index documents for retrieval-augmented answers.
- Build agent workflows using datasets and prebuilt templates.
Who it is for
- Teams building document-grounded assistants and knowledge pipelines.
Requirements
- Docker deployment recommends 4 CPU cores, 16 GB RAM, and 50 GB disk as a starting point.
- Configure the document engine, storage services, and model/embedding access.
Limitations
- Resource requirements scale with parsing jobs, document volume, and concurrency.
- Retrieval and citations still require evaluation against the underlying documents.
Cost and privacy notes
- Parsing, storage, model calls, and hosting can each contribute to cost.
- Review document retention and external model/embedding routing before ingesting confidential files.
Illustrative example (untested)
Illustrative, not tested: index a small document set and verify an answer against the retrieved passages.
License
Apache-2.0; reviewed the repository LICENSE.