I'm excited to share Local LLM Helper with you today! // What inspired me: As someone who lives in Obsidian for notes and knowledge management, I was frustrated by the disconnect between my notes and AI tools. Every time I wanted to process text or search my vault semantically, I had to copy-paste into separate apps, send my private notes to cloud APIs, or deal with clunky workflows. I wanted the power of LLMs directly in my note-taking environment without compromising privacy. // The problem I was solving: Most AI-powered note tools force you to choose between convenience and privacy. They either lock you into proprietary cloud services or don't give you the flexibility to use different models. // I wanted something that: Works with local LLMs (Ollama, LM Studio) to keep data private Supports any OpenAI-compatible provider for flexibility Integrates seamlessly into Obsidian's workflow Makes semantic search and RAG accessible to everyone // How I approached it: I started with basic text transformation commands (summarize, rephrase) in v1.0, then progressively added features based on what I personally needed: chat interfaces, personas, and eventually full RAG capabilities. The big breakthrough came in v2.3 when I redesigned the chunking strategy with overlap and added incremental indexing. This made the RAG chat actually useful for large vaults without constant re-indexing. I'm really proud of how this has evolved from a simple text processor to a complete local AI toolkit for Obsidian. Would love to hear your thoughts and what features you'd find most useful!

Obsidian Local LLM Helper

Private AI chat & semantic search for Obsidian notes

Private AI chat & semantic search for Obsidian notes