GH Repository · Zleap-AI
SAG
A new SOTA for RAG — an original retrieval architecture and an open-source knowledge base for humans and agents.
- stars
- 2,495
- 30-day movement
- +145/day
- Related entries
- 60
- Connections
- 1
aiknowledge-graphllmdata-engineeringraggraphragagentknowledge-graphsmakeknowledge-basedockervector-searchsagPython
SAG is a Python retrieval-augmented generation (RAG) system from Zleap AI, described as combining an original retrieval architecture with an open-source knowledge base intended for both humans and agents. Its topics indicate knowledge-graph and vector-search components, with a GraphRAG orientation.
Reach for it when you want a graph-aware RAG stack you can self-host and adapt under an MIT licence.
Use it to
- Build a knowledge-graph-backed RAG pipeline
- Serve a knowledge base to LLM agents
- Run retrieval with vector search plus graph traversal
- Deploy the stack via the provided Docker toolchain
For Engineers building RAG and agent retrieval systems
- Role
- rag
- Language
- Python
- Licence
- MIT
- Forks
- 162
- Open issues
- 2
- Last push
- 2026-09-17
- Latest release
- v1.3.0 · 2026-07-22
topicsragknowledge-graphvector-searchllmgraphragagents