SemantiNote — local-first smart notes
Free & 100% local. Take notes, then find them by meaning and chat with a local AI about them — nothing ever leaves your computer.
SemantiNote is a note-taking app inside VS Code. Take your meeting notes (or anything else), then talk to a local open-source LLM — downloaded with Ollama — about them: ask questions, get summaries, add information, and organize. Find any note instantly by meaning (semantic search) or by keyword, and let the AI web search and even edit your notes for you. No cloud, no subscriptions, no coding required.
💙 SemantiNote is free thanks to support from people like you. If it saves you time, consider sponsoring the developer — it keeps the project alive and independent.
What SemantiNote can do
📝 Notes, organized simply
- Pick a notebook folder, then create folders and notes right in the sidebar.
- Drag & drop to reorganize — drop a note on a folder to move it in, or on empty space to move it to the top level. Right-click for Rename / Move to… / Delete.
- Notes are plain Markdown files on your machine — portable, backup-friendly, and yours.
🔎 Find anything — no digging through folders and files
- Semantic search finds notes by meaning, powered by a local embedding model. Ask "what did we decide about the budget?" and it surfaces the right meeting note even if it never used those exact words — so you don't have to remember which folder or file it's in.
- Keyword search finds notes containing your exact words and highlights every match.
- A status indicator always shows whether your notes are indexed and which embedding model did it — and warns you if they aren't searchable yet.
💬 Chat with a local LLM about your notes
- Click ✨ Ask SemantiNote (in the note's toolbar or the status bar) — no special syntax to learn.
- Ask questions, get summaries, add information, and organize your notes in plain language. It uses the note(s) you have open plus semantic search across your whole notebook, and cites the notes it used.
- It's a tool-using agent: the model decides when to search your notes, read a note, search the web, or create/edit a note for you — every edit is shown as a red/green diff you Keep or Undo.
- Every answer shows which model and mode it's using, so you always know what's running.
🎤 Voice-to-text — speak your notes
- Open a note, then click the 🎤 mic button at the top-right of the note's editor ("Dictate into Note"), talk, and your words are typed at the cursor. Click the ⏹ Stop button next to it (or press Escape) when you're done.
- One-time setup: install Microsoft's free VS Code Speech extension (on-device, works offline), then grant microphone permission and restart VS Code:
- Windows: Settings → Privacy & security → Microphone → turn on Microphone access and Let desktop apps access your microphone.
- Mac: System Settings → Privacy & Security → Microphone → turn on Visual Studio Code.
- If nothing appears when you talk, it's almost always the microphone permission above (dictation fails silently without it), or a one-time speech-model download finishing.
🧠 Pick & manage your models
- A Models panel lists your installed Ollama models with capability badges (tools ✓ / embedding / vision). Pull new models or remove them right there.
- A 🔑 key button saves your Ollama API key — which enables Ollama's hosted web search (otherwise the chat uses DuckDuckGo, no key needed).
🔒 Private and free
- Everything runs locally through Ollama — zero telemetry, zero API costs, no cloud account, and no data ever leaves your computer. The only feature that touches the internet is optional web search.
Requirements
- Visual Studio Code 1.104 or newer.
- Python 3.10+ on your PATH — download (on Windows tick "Add python.exe to PATH").
- Ollama installed and running — download.
- An embedding model for semantic search:
ollama pull nomic-embed-text (any embedding-capable model works).
- A chat model that supports tools, e.g.
ollama pull qwen3:4b.
On first launch SemantiNote creates its own small Python environment and installs its backend — a one-time setup that takes a few minutes. Later launches are fast.
Getting Started
- Install Python, Ollama, an embedding model, and a chat model — the Set Up SemantiNote walkthrough opens automatically and guides you.
- Click the SemantiNote icon (the feather) in the Activity Bar and Set Notebook Folder.
- Create a note, then click ✨ Ask SemantiNote to chat about it.
Privacy
SemantiNote is local-first and collects no usage data. Your notes, prompts, and generated content are processed by your local Ollama server unless you explicitly use web search. Any Ollama API key you enter is stored in VS Code SecretStorage.
👋 Want to bring AI to your business? Let's talk.
SemantiNote is built by an independent AI consultant who helps companies and teams actually adopt AI — not just talk about it. If your business wants to put AI to work — building custom tools and agents, automating workflows, training your team, or just figuring out where to start — I can work with your company directly, hands-on.
👉 Get in touch — let's talk about what AI can do for your business.
Feedback, feature requests & bug reports
Got a feature idea, found a bug, or have a question? Same place:
Your feedback shapes what gets built next — every request is read.
Powered by Ollama.
| |
