Skip to content
Get packs

Leon

Open-source personal AI assistant that can run tools, keep memory, and work locally.

GitHub Download source
Stars
17.6k
Forks
1.5k
Licence
MIT
Language
TypeScript
Last update
4 days ago
Type
App

What it is

Leon is an open-source personal AI assistant that runs on your own computer and can use tools to carry out tasks rather than only replying with text. The README describes it as built around tools, context, memory, and agentic execution, with support for local and remote AI providers. The project is currently focused on a 2.0 Developer Preview on the develop branch, and the README notes that new documentation is not ready yet.

  • Run Leon locally and open its web interface at localhost:5366
  • Use skills for search, productivity, system utilities, media workflows, and coding assistance, as listed in the README
  • Explore voice and audio features included in the repository
  • Study how skills, actions, tools, and bridges are structured

Great for

  • Developers who want to inspect or extend an open-source assistant
  • People who want an assistant that can run locally for privacy reasons
  • Contributors interested in the 2.0 Developer Preview architecture
  • Users comfortable with a preview that may change between versions

Not ideal for

  • People who want a polished, stable assistant out of the box: the README says the 2.0 docs are not ready and the legacy master branch is the more stable version
  • Non-technical users who need a simple download-and-use app: setup requires Node.js and pnpm from the command line
  • Anyone needing finished, complete documentation: the README says the public docs site may lag behind the new core
Step-by-step guide: how to use LeonInstall, first run, examples, tips and FAQs
Leon alternativesLeon vs OpenClawLeon vs OpenWhispr

How to get it

  1. 1

    Clone and build from source

    You'll need: Node.js >= 24.0.0 · pnpm (installed via npm) · Supported OS: Linux, macOS, or Windows

    Clone the repository

    git clone https://github.com/leon-ai/leon.git

    Go to the project root

    cd leon

    Install pnpm

    npm install --global pnpm@latest

    Install dependencies

    pnpm install
  2. 2

    Run it

    • Run pnpm start to launch Leon
    • Open http://localhost:5366 in your browser, which the README says is the default address
    • Run pnpm run check to confirm the setup went well
  3. 3

    Get the most out of it

    • Read core/context/LEON.md and core/context/ARCHITECTURE.md for the most accurate description of current behavior
    • Try the different run modes described in the README: smart, controlled, and agent
    • Add agent skills backed by SKILL.md files or native skills under skills/
    • Join the Discord or follow the roadmap to track 2.0 progress

Downloads

Source code (.zip)develop branch · straight from GitHubOfficial website & docsgetleon.ai

Strengths

  • Open source under the MIT license, so the code can be inspected and modified
  • Supports local AI models as well as remote providers, per the README
  • Organized around explicit tools and layered memory rather than free-form guessing alone

Trade-offs

  • The README states this is a developer preview in transition, so behavior and docs may change
  • Contribution is currently limited to progressively onboarded contributors, and development is largely done in spare time
  • Setup requires a command-line workflow and Node.js 24 or newer

How it compares

ProjectStarsTypeLevelUpdatedBest for
leon (this)17.6kAppintermediate4 days agoDevelopers who want to inspect or extend an open-source assistant
piper11.3kAppintermediate1.2 years agoDevelopers who want offline speech output in a local app or home automation setup
whisper.cpp54.9kLibraryintermediate7 days agoDevelopers who want to embed offline speech-to-text in a C, C++, or other native app
faster-whisper26.4kLibraryintermediate6 days agoPython developers who need to add speech-to-text to an application
FunASR20.9kLibraryintermediate4 days agoDevelopers who want to transcribe audio in Python with a pretrained model
Amphion10.3kLibraryadvanced8 months agoResearchers who want reproducible implementations of speech and audio generation models

Stats synced from GitHub. The assessment above was drafted with AI from the project's README on 2026-10-29 and may miss recent changes — check the README before relying on a command. MIT licence: free to use, modify and distribute, including commercially, as long as you keep the copyright notice.