QwenPaw
Personal AI assistant supporting multiple chat apps, local runtimes, and extensible plugins.
Deploy locally or in the cloud, extend with skills, and connect across every chat channel.
QwenPaw is a personal AI assistant framework designed for multi-channel communication and agent orchestration. It suits professionals looking to self-host custom AI workflows across platforms like Discord, Telegram, and DingTalk.
What it is
The tool provides a three-layer memory architecture via ReMe, kernel-level sandboxing for security, and multi-agent parallel execution. It supports local runtimes like Ollama and QwenPaw-Flash alongside 14+ cloud providers.
Who it's for
It suits developers and system administrators who want to deploy a locally-controlled agent ecosystem with custom workflows. It does not suit non-technical users who require a zero-setup hosted consumer app without configuration.
Key features
Three-Layer Memory
Maintains live working context, verbatim history, and a self-evolving personal Markdown knowledge base.
Security Sandbox
Features kernel-level sandboxing, tool guards, and file guards to block dangerous commands before they execute.
Multi-Channel Ops
Connects a single instance simultaneously to DingTalk, Lark, WeChat, Discord, Telegram, and other chat apps.
File Workspace
Offers unified file navigation, previews, editing, diffs, and uploads across project and agent files.
Up and running in 3 steps
Our honest verdict
What's great
- Supports local execution via QwenPaw-Flash or Ollama without requiring a cloud API key
- Connects a single instance across multiple chat applications like Telegram, Discord, and Lark
- Includes built-in kernel-level sandboxing and tool guards to restrict dangerous commands
- Features a three-layer memory system that turns conversations into searchable Markdown files
Where it falls short
- Requires technical skill to set up, configure, and maintain across self-hosted environments
- Running local models or multiple agent loops demands adequate local hardware resources
- Managing multi-channel credentials and agent permissions requires ongoing administrative oversight
Pricing & license
Free under its open-source license, with the only cost being the hardware or cloud infrastructure you use to run it.
Open remote AI-training roles
874 open remote AI-training roles, updated November 15, 2026; most pay $20–$170 an hour. Zekai lists them free; the application goes to the employer’s platform.
Alternatives
These reviewed AI tools do the same job instantly — from our scored directory.
Use it, or have us set it up
FAQ
Can I run QwenPaw without a cloud API key?
Yes, you can run it using the built-in QwenPaw Local runtime or connect it to local tools like Ollama and LM Studio.
Which chat applications are supported?
QwenPaw supports DingTalk, Lark, WeChat, Discord, Telegram, iMessage, and QQ from a single instance.
