OpenSource-Hub
O

OpenHuman

SHA-256
3.1k stars·AI Productivity·SHA-256 checksum verified

Personal AI super intelligence that knows you in minutes. Open-source, private, simple desktop agent with memory tree, 118+ integrations, and a cute mascot.

Smart Download

Download Download Version

v0.53.22 · 185.9 MB

Open-source desktop AI assistant that learns your workflow in minutes, with persistent memory and one-click integrations.

Core Features

  • Desktop mascot that speaks, reacts, joins Google Meets, and thinks in background
  • One-click OAuth to 118+ services (Gmail, Notion, GitHub, Slack, etc.) with auto-fetch every 20 minutes
  • Local Memory Tree + Obsidian Wiki: compresses data into Markdown files you can browse and edit
  • Smart Token Compression (TokenJuice) reduces cost & latency up to 80%
  • Built-in voice (STT/TTS with lip-sync), web search, full coder toolset (filesystem, git, lint, test)

What It Can't Do

  • •Early Beta: expect rough edges and missing features. 2. Model routing requires a subscription; local AI needs Ollama installed. 3. Auto-fetch runs every 20 minutes; keep the app running. 4. Memory tree and wiki files are stored locally – back up ~/.openhuman if needed.

Use Cases

  • Personal knowledge management: sync emails, docs, calendar into a searchable local wiki
  • Workflow automation: let AI handle email, task management, code generation
  • Meeting assistant: mascot joins Google Meet for real-time transcription, summarization, reminders

Detailed Introduction

OpenHuman is an open-source agentic assistant that integrates with your daily life out of the box. It provides a clean desktop UI with a speaking mascot that joins Google Meets, remembers you across weeks, and thinks in the background. With one-click OAuth, it connects to 118+ third-party services (Gmail, Notion, GitHub, Slack, etc.) and auto-fetches fresh data every 20 minutes into a local-first Memory Tree and Obsidian wiki. It includes native voice (STT/TTS with lip-sync), web search, a full coder toolset, and smart token compression (TokenJuice) that reduces costs and latency by up to 80%. Optional local AI via Ollama. Privacy-first: workflow data stays on-device, encrypted locally. Compared to Claude Cowork (proprietary, chat-scoped memory) or OpenClaw (terminal-first, plugin-reliant), OpenHuman offers a minutes-to-value experience with persistent memory and a unified subscription.

Troubleshooting & FAQ (1)

Troubleshooting
Why does the test 'pipeline_status_reports_chunk_aggregates_after_ingest' fail intermittently in CI?

This is a known flaky test caused by a race condition or ordering dependency in openhuman's memory_tree. Temporary workaround: mark the test with #[ignore] and a TODO comment, or add --retry 1 to your CI command. A permanent fix is under investigation at issue #3256.

GitHub Issue #3256

Tags

AI assistantdesktop agentmemory treeobsidianopen sourcevoiceintegrations

Getting Started

1

Download installer

Click the button above to download the installer for your system

Windows· 185.9 MBmacOS· 168.1 MBLinux· 183.2 MB
2

Install the software

Double-click the downloaded installer and follow the prompts

3

Step 1: Download DMG/EXE from https://tinyhumans.ai/openhuman or run one-liner install (macOS/Linux: curl -fsSL https://raw.githubusercontent.com/tinyhumansai/openhuman/main/scripts/install.sh | bash; Windows: irm ... | iex)

4

Step 2: Run the installer and follow the wizard

5

Step 3: Launch OpenHuman, connect your accounts (Gmail, Notion, GitHub, etc.), wait for auto-fetch, and start using

Install Guide
  1. Step 1: Download DMG/EXE from https://tinyhumans.ai/openhuman or run one-liner install (macOS/Linux: curl -fsSL https://raw.githubusercontent.com/tinyhumansai/openhuman/main/scripts/install.sh | bash; Windows: irm ... | iex)
  2. Step 2: Run the installer and follow the wizard
  3. Step 3: Launch OpenHuman, connect your accounts (Gmail, Notion, GitHub, etc.), wait for auto-fetch, and start using
File Integrity

SHA-256 checksum verified

Checksum extracted from GitHub official Release page

SHA256 Checksum

9942bed5cb5e5cea3941e6eaed94a589c7e6ca17b2ec7bc3bef06cfc1bbbfeeb

This checksum is extracted from the GitHub Release page. Verify file integrity after download.

All SHA-256 checksums on this platform are extracted from the project's official GitHub Release page, without any modification. You can independently verify them on the GitHub Releases page.

Open Source Transparency

View GitHub Source
Environment Guide

Uninstall Info

Delete the OpenHuman app bundle/executable and remove ~/.openhuman directory (contains local memory and config).

No Extra Dependencies

Ready to use after download. No additional runtime required.

Project Info
LicenseGNU
Last Updated2026-06-27 12:30:23
GitHub RepositoryOfficial Website

Having issues? Check the FAQ below

1 FAQ

Similar Projects

G

goose

Goose is a general-purpose AI agent that runs on your machine. Use it for coding, research, writing, automation, and more. It offers a native desktop app (macOS, Linux, Windows), a full CLI, and an API. Works with 15+ providers (Anthropic, OpenAI, Google, Ollama, etc.) and connects to 70+ extensions via the Model Context Protocol. Built in Rust, part of the Linux Foundation.

D

daily_stock_analysis

An open-source AI stock analysis system for A/H/US markets that generates daily decision dashboards and pushes them to WeChat Work, Feishu, Telegram, Discord, Slack, or email. Deploy via GitHub Actions for free.

O

ollama

Ollama lets you download, run, and manage large language models locally. One command, multiple platforms, endless possibilities.