An open-source RAG engine combining document parsing, knowledge retrieval and agents to supply traceable context for model applications.
Core features
- Parse and retrieve information from multiple document formats
- Combine vector, keyword and reranking retrieval
- Build knowledge Q&A and multi-step retrieval workflows
When to use it
Build organizational knowledge bases, document Q&A and retrieval applications that need source references.
Plans and practical limits
The official repository lists Apache-2.0. Self-hosting infrastructure, model calls and commercial cloud services have separate costs.
Traceable references do not guarantee correct answers. Check deployment resources and model configuration in the official documentation.
Sources and review
Official source 1 · Official source 2 · Checked 19 October 2026. Based on public official sources; signed-in and paid features were not tested.
