English | 中文

AgentScope

Start building LLM-empowered multi-agent applications in an easier way.

If you find our work helpful, please kindly cite our paper.

Welcome to join our community on

Discord	DingTalk

News

• [2024-04-27] AgentScope Workstation is now online! You are welcome to try building your multi-agent application simply with our drag-and-drop platform and ask our copilot questions about AgentScope!

• [2024-04-19] AgentScope supports Llama3 now! We provide scripts and example model configuration for quick set-up. Feel free to try llama3 in our examples!

• [2024-04-06] We release AgentScope v0.0.3 now!

• [2024-04-06] New examples Gomoku, Conversation with ReAct Agent, Conversation with RAG Agent and Distributed Parallel Search are available now!

• [2024-03-19] We release AgentScope v0.0.2 now! In this new version, AgentScope supports ollama(A local CPU inference engine), DashScope and Google Gemini APIs.

• [2024-03-19] New examples "Autonomous Conversation with Mentions" and "Basic Conversation with LangChain library" are available now!

• [2024-03-19] The Chinese tutorial of AgentScope is online now!

• [2024-02-27] We release AgentScope v0.0.1 now, which is also available in PyPI!

• [2024-02-14] We release our paper "AgentScope: A Flexible yet Robust Multi-Agent Platform" in arXiv now!

What's AgentScope?

AgentScope is an innovative multi-agent platform designed to empower developers to build multi-agent applications with large-scale models. It features three high-level capabilities:

• 🤝 Easy-to-Use: Designed for developers, with fruitful components, comprehensive documentation, and broad compatibility. Besides, AgentScope Workstation provides a drag-and-drop programming platform and a copilot for beginners of AgentScope!

• ✅ High Robustness: Supporting customized fault-tolerance controls and retry mechanisms to enhance application stability.

• 🚀 Actor-Based Distribution: Building distributed multi-agent applications in a centralized programming manner for streamlined development.

Supported Model Libraries

AgentScope provides a list of ModelWrapper to support both local model services and third-party model APIs.

Supported Local Model Deployment

AgentScope enables developers to rapidly deploy local model services using the following libraries.

• ollama (CPU inference)
• Flask + Transformers
• Flask + ModelScope
• FastChat
• vllm

Supported Services

• Web Search
• Data Query
• Retrieval
• Code Execution
• File Operation
• Text Processing

Example Applications

• Model

  • Using Llama3 in AgentScope

• Conversation

  • Basic Conversation
  • Autonomous Conversation with Mentions
  • Self-Organizing Conversation
  • Basic Conversation with LangChain library
  • Conversation with ReAct Agent
  • Conversation with RAG Agent

• Game

  • Gomoku
  • Werewolf

• Distribution

  • Distributed Conversation
  • Distributed Debate
  • Distributed Parallel Search
  • Distributed Large Scale Simulation

More models, services and examples are coming soon!

Installation

AgentScope requires Python 3.9 or higher.

Note: This project is currently in active development, it's recommended to install AgentScope from source.

From source

• Install AgentScope in editable mode:

# Pull the source code from GitHub
git clone https://github.com/modelscope/agentscope.git

# Install the package in editable mode
cd agentscope
pip install -e .

• To build distributed multi-agent applications:

# On windows
pip install -e .[distribute]
# On mac
pip install -e .\[distribute\]

Using pip

• Install AgentScope from pip:

pip install agentscope

Quick Start

Configuration

In AgentScope, the model deployment and invocation are decoupled by ModelWrapper.

To use these model wrappers, you need to prepare a model config file as follows.

model_config = {
    # The identifies of your config and used model wrapper
    "config_name": "{your_config_name}",          # The name to identify the config
    "model_type": "{model_type}",                 # The type to identify the model wrapper

    # Detailed parameters into initialize the model wrapper
    # ...
}

Taking OpenAI Chat API as an example, the model configuration is as follows:

openai_model_config = {
    "config_name": "my_openai_config",             # The name to identify the config
    "model_type": "openai_chat",                   # The type to identify the model wrapper

    # Detailed parameters into initialize the model wrapper
    "model_name": "gpt-4",                         # The used model in openai API, e.g. gpt-4, gpt-3.5-turbo, etc.
    "api_key": "xxx",                              # The API key for OpenAI API. If not set, env
                                                   # variable OPENAI_API_KEY will be used.
    "organization": "xxx",                         # The organization for OpenAI API. If not set, env
                                                   # variable OPENAI_ORGANIZATION will be used.
}

More details about how to set up local model services and prepare model configurations is in our tutorial.

Create Agents

Create built-in user and assistant agents as follows.

from agentscope.agents import DialogAgent, UserAgent
import agentscope

# Load model configs
agentscope.init(model_configs="./model_configs.json")

# Create a dialog agent and a user agent
dialog_agent = DialogAgent(name="assistant",
                           model_config_name="my_openai_config")
user_agent = UserAgent()

Construct Conversation

In AgentScope, message is the bridge among agents, which is a dict that contains two necessary fields name and content and an optional field url to local files (image, video or audio) or website.

from agentscope.message import Msg

x = Msg(name="Alice", content="Hi!")
x = Msg("Bob", "What about this picture I took?", url="/path/to/picture.jpg")

Start a conversation between two agents (e.g. dialog_agent and user_agent) with the following code:

x = None
while True:
    x = dialog_agent(x)
    x = user_agent(x)
    if x.content == "exit":  # user input "exit" to exit the conversation_basic
        break

AgentScope Studio

AgentScope provides an easy-to-use runtime user interface capable of displaying multimodal output on the front end, including text, images, audio and video. To start a studio, you should install the full version of AgentScope.

# On windows
pip install -e .[full]
# On mac
pip install -e .\[full\]

Once installed, you can just run

as_studio  path/to/your/script.py

Then the studio will be launched at localhost:xxxx, and you can see the UI similar to the following: To be able to use the as_studio functionality, please implement the main function in your code. More detail can be found in src/agentscope/web/README.md.

Tutorial

• Getting Started
  • About AgentScope
  • Installation
  • Quick Start
  • Crafting Your First Application
  • Logging and WebUI
• Advanced Exploration
  • Customize Your Own Agent
  • Pipeline and MsgHub
  • Model
  • Service
  • Memory
  • Prompt Engineering
  • Monitor
  • Distribution
• Get Involved
  • Join AgentScope Community
  • Contribute to AgentScope

License

AgentScope is released under Apache License 2.0.

Contributing

Contributions are always welcomed!

We provide a developer version with additional pre-commit hooks to perform checks compared to the official version:

# For windows
pip install -e .[dev]
# For mac
pip install -e .\[dev\]

# Install pre-commit hooks
pre-commit install

Please refer to our Contribution Guide for more details.

References

If you find our work helpful for your research or application, please cite our paper:

@article{agentscope,
  author  = {Dawei Gao and
             Zitao Li and
             Weirui Kuang and
             Xuchen Pan and
             Daoyuan Chen and
             Zhijian Ma and
             Bingchen Qian and
             Liuyi Yao and
             Lin Zhu and
             Chen Cheng and
             Hongzhu Shi and
             Yaliang Li and
             Bolin Ding and
             Jingren Zhou},
  title   = {AgentScope: A Flexible yet Robust Multi-Agent Platform},
  journal = {CoRR},
  volume  = {abs/2402.14034},
  year    = {2024},
}