Skip to content

Latest commit

 

History

History
100 lines (68 loc) · 2.16 KB

README.en.md

File metadata and controls

100 lines (68 loc) · 2.16 KB

LmNotebook

中文 | English

LmNotebook is a local-first AI notebook for individuals and small teams. It is built with Electron + Vue 3 + FastAPI + LangGraph, with notes, vector index, and chat session data stored locally by default.

Why LmNotebook

  • Local-first: your data stays on your machine by default
  • Model-flexible: multi-provider and multi-model switching
  • Agent-ready: Ask / Agent modes with approval workflow
  • Release-ready: Windows / macOS / Linux distribution pipeline

Core Capabilities

Notes

  • Markdown editing with structured formatting (headings, lists, code, tables, math)
  • Folders / categories / pin / trash
  • Full-text search and quick filtering
  • Auto-save and local backup

AI Agent

  • LangGraph-based orchestration
  • Tool calls: read, search, create, update, rename, categorize, delete
  • Approval workflow: manual review / auto-accept
  • Session modes: Ask (read-only) / Agent (action-enabled)
  • Attachments: image and file upload, paste, drag-and-drop

RAG

  • Local FAISS vector retrieval
  • Incremental note vector sync
  • Configurable embedding model

Tech Stack

  • Frontend: Electron + Vue 3 + TypeScript + Pinia + TipTap
  • Backend: Python + FastAPI + LangGraph + LangChain
  • Retrieval: FAISS
  • Storage: SQLite

Quick Start

Requirements

  • Node.js 18+
  • Python 3.10+
  • npm or pnpm

Install

npm install
cd src/backend
pip install -r requirements.txt

Configure Models

Copy src/backend/.env.example to src/backend/.env:

OPENAI_BASE_URL=https://api.openai.com/v1
OPENAI_API_KEY=your_api_key
MODEL_NAME=gpt-4o-mini

DASHSCOPE_API_KEY=your_dashscope_key
EMBEDDING_MODEL=text-embedding-v3

You can also manage providers and models from in-app Model Settings. UI config is persisted in the user data directory and overrides .env defaults.

Run (Dev)

# backend
cd src/backend
python main.py

# frontend (new terminal)
npm run dev

Packaging and Release

  • Packaging guide: PACKAGING.md
  • Release checklist: docs/release-checklist.md

Docs

  • Backend docs: src/backend/README.md
  • Dev notes: docs/dev/
  • Plans and design docs: docs/plans/

License

MIT