Kwipu
Ask an Obsidian vault questions and get cited answers: a Python graph RAG on Ollama with an MCP server and a 3D view, though its default LLM runs in the cloud.
// RAG
6 entries tagged with "rag"
// ENTRIES
6 entries
LLM Wiki
Compile documents once into a Markdown wiki an LLM keeps current, rather than re-reading them for every question. GPL-3.0 desktop app with graph and MCP server.
Hindsight
Long-term memory for AI agents: an LLM turns writes into facts and entities, four retrievers bring them back. MIT server on PostgreSQL, 1.0 since October 2026.
PaddleOCR
Pull text, tables and formulas out of PDFs and images as Markdown or JSON for an LLM: OCR and vision models under Apache 2.0, on PaddlePaddle by default.
Flowise
Drag-and-drop AI agents and RAG chains, each served over its own REST API. The project was archived in August 2026 and ships no releases or security fixes.
Dify
LLM chatbots, agents and RAG workflows built in a web console, each with its own API. Its modified Apache 2.0 license requires permission for multi-tenant use.