Qwen-Agent is a Python framework for developing LLM applications using Qwen models, featuring function calling, Model Context Protocol (MCP) integration, retrieval-augmented generation (RAG), and code interpretation. It requires writing Python code to instantiate agents and involves provisioning an external model service such as DashScope or connecting to a local runtime.
Project overview
Provides a built-in RAG solution for documents involving up to 1M tokens and native support for parallel function calls for Qwen models.
Project type
MCP · AI Agent · RAG
Use cases
Knowledge Q&A · Coding & Development · Automation
Deployment
Refer to project documentation
License
Apache-2.0
Best for
Developers and AI engineers building LLM applications who can write Python code to instantiate agents.
Key capabilities
The LLM classes provide function calling, including native support for Parallel Function Calls.
Provides the ability to select required tools from the open-source MCP server website and configure the relevant environment.
Provides a fast RAG solution for doing question-answering over super-long documents involving 1M tokens.
Includes a browser assistant built upon Qwen-Agent.
Limitations and risks
The Docker container-based code interpreter implements basic sandbox isolation, but it should still be used with caution in production environments.
The python executor is not sandboxed and is intended for local testing only, not for production use.
Getting started
Install the package via pip. To use the web GUI, Python 3.10 or higher is required. Configure the LLM by either setting the DASHSCOPE_API_KEY environment variable to connect to the DashScope model service or provisioning a local deployment. Once configured, create and run the Assistant agent in Python.
Evidence and sources
GitHub project description: Agent framework and applications built upon Qwen>=3.0, featuring Function Calling, MCP, Code Interpreter, RAG, Chrome extension, etc.
README: Qwen-Agent is a framework for developing LLM applications based on the instruction following, tool usage, planning, and memory capabilities of Qwen.
README: Now Qwen-Agent plays as the backend of [Qwen Chat](https://chat.qwen.ai/).
README: 🔥🔥🔥Feb 16, 2026: Open-sourced Qwen3.5. For usage examples, refer to [Qwen3.5 Agent Demo](./examples/assistant_qwen3.5.py).
README: Yes. The LLM classes provide [function calling](https://github.com/QwenLM/Qwen-Agent/blob/main/examples/function_calling.py). Additionally, some Agent classes also are built upon…