Jev in production › Search, ranking and matching

SAG Wiki

Local knowledge base app where Jev judges BM25 candidates as evidence for retrieval. 80.68% Recall@5 on MuSiQue, 15.55 points above HippoRAG 2 (author).

Open on GitHub ↗

80.68%measured against a baseline, as published by the source
Use
Search, ranking and matching
Industry
Education and research
Form
Open-source tool
Stage
Announced
Listed
2026-11-10
Found via
x
Repository
Zleap-AI/SAG-Wiki
Stars
1,446
Forks
168
Last push
2026-11-18
Language
Rust
License
MIT
SAG Wiki screenshot
docs/assets/readme/sag2-retrieval-comparison-en.png in the Zleap-AI/SAG-Wiki README, MIT; shown from GitHub.

The README opens with

A lightweight knowledge base for people and agents. Index and search your local files without LLM calls, then ask questions with answers you can trace to the source.

https://github.com/user-attachments/assets/e9c5499d-6057-4add-ac7e-f57eae381a2f

PDF basic text can become searchable before background OCR and table enhancement finish. Independent basic indexing keeps supported local Office/RTF documents moving, while PDF segments reuse loaded models. See the changelog for release-specific changes and the bilingual documentation for usage. The download link below points to the latest published stable Release.

Badge

For the project's own README, linking back here:

Listed in Jev in production

Also used for search, ranking and matching