masterFoad/NanoSage
Local LLM Powered Recursive Search & Smart Knowledge Explorer
Implements Monte Carlo-driven recursive search with branching subqueries, guided by relevance scoring to balance exploration depth. Combines local document retrieval with Tavily's academic search API, using a hybrid embedding system where SigLIP/CLIP handles vision content (PDFs, images) and all-MiniLM processes text, ensuring consistent 384D embeddings across content types. Generates structured Markdown reports with a Table of Contents tracking the search graph, powered by Ollama's Gemma 2B for RAG-based summarization.
261 stars.
Stars
261
Forks
33
Language
Python
License
Apache-2.0
Category
Last pushed
Oct 18, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/rag/masterFoad/NanoSage"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related tools
LeDat98/NexusRAG
Hybrid RAG system combining vector search, knowledge graph (LightRAG), and cross-encoder...
Zakk-Yang/nexusync
Intuitive RAG system on top of LllamaIndex
Serhii2009/brix-protocol
Runtime Reliability Infrastructure for LLM Pipelines — enforce deterministic rules, measure the...
Ambuj123-lab/agentic-rag-financial-parser
Enterprise RAG ecosystem managing 15,000+ semantic chunks. Features hybrid parsing...
huangjia2019/rag-project03-audit
[RAG训练营] u.geekbang.org/subject/airag/1009927 ESG合规审计系统 - 可持续发展报告检查工具