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.

• Visit our workstation to build multi-agent applications with dragging-and-dropping.

• Welcome to join our community on

Discord	DingTalk

---

News

• [2024-07-18] AgentScope supports streaming mode now! Refer to our tutorial and example conversation in stream mode for more details.

• [2024-07-15] AgentScope has implemented the Mixture-of-Agents algorithm. Refer to our MoA example for more details.

• [2024-06-14] A new prompt tuning module is available in AgentScope to help developers generate and optimize the agents' system prompts! Refer to our tutorial for more details!

• [2024-06-11] The RAG functionality is available for agents in AgentScope now! A quick introduction to RAG in AgentScope can help you equip your agent with external knowledge!

• [2024-06-09] We release AgentScope v0.0.5 now! In this new version, AgentScope Workstation (the online version is running on agentscope.io) is open-sourced with the refactored AgentScope Studio!

• [2024-05-24] We are pleased to announce that features related to the AgentScope Workstation will soon be open-sourced! The online website services are temporarily offline. The online website service will be upgraded and back online shortly. Stay tuned...

• [2024-05-15] A new Parser Module for formatted response is added in AgentScope! Refer to our tutorial for more details. The DictDialogAgent and werewolf game example are updated simultaneously.

Werewolf-running-example.mp4

• [2024-05-14] Dear AgentScope users, we are conducting a survey on AgentScope Workstation & Copilot user experience. We currently need your valuable feedback to help us improve the experience of AgentScope's Drag & Drop multi-agent application development and Copilot. Your feedback is valuable and the survey will take about 3~5 minutes. Please click URL to participate in questionnaire surveys. Thank you very much for your support and contribution!

• [2024-05-14] AgentScope supports gpt-4o as well as other OpenAI vision models now! Try gpt-4o with its model configuration and new example Conversation with gpt-4o!

• [2024-04-30] We release AgentScope v0.0.4 now!

• [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 Optimization 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.

API	Task	Model Wrapper	Configuration	Some Supported Models
OpenAI API	Chat	OpenAIChatWrapper	guidance template	gpt-4o, gpt-4, gpt-3.5-turbo, ...
	Embedding	OpenAIEmbeddingWrapper	guidance template	text-embedding-ada-002, ...
	DALL·E	OpenAIDALLEWrapper	guidance template	dall-e-2, dall-e-3
DashScope API	Chat	DashScopeChatWrapper	guidance template	qwen-plus, qwen-max, ...
	Image Synthesis	DashScopeImageSynthesisWrapper	guidance template	wanx-v1
	Text Embedding	DashScopeTextEmbeddingWrapper	guidance template	text-embedding-v1, text-embedding-v2, ...
	Multimodal	DashScopeMultiModalWrapper	guidance template	qwen-vl-max, qwen-vl-chat-v1, qwen-audio-chat
Gemini API	Chat	GeminiChatWrapper	guidance template	gemini-pro, ...
	Embedding	GeminiEmbeddingWrapper	guidance template	models/embedding-001, ...
ZhipuAI API	Chat	ZhipuAIChatWrapper	guidance template	glm-4, ...
	Embedding	ZhipuAIEmbeddingWrapper	guidance template	embedding-2, ...
ollama	Chat	OllamaChatWrapper	guidance template	llama3, llama2, Mistral, ...
	Embedding	OllamaEmbeddingWrapper	guidance template	llama2, Mistral, ...
	Generation	OllamaGenerationWrapper	guidance template	llama2, Mistral, ...
LiteLLM API	Chat	LiteLLMChatWrapper	guidance template	models supported by litellm...
Yi API	Chat	YiChatWrapper	guidance template	yi-large, yi-medium, ...
Post Request based API	-	PostAPIModelWrapper	guidance template	-

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
• Multi Modality
• Wikipedia Search and Retrieval
• TripAdvisor Search

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 in Natural Language to Query SQL
  • Conversation with RAG Agent
  • Conversation with gpt-4o
  • Conversation with Software Engineering Agent
  • Conversation with Customized Tools
  • Mixture of Agents Algorithm
  • Conversation in Stream Mode
  • Conversation with CodeAct Agent
  • Conversation with Router Agent

• Game

  • Gomoku
  • Werewolf

• Distribution

  • Distributed Conversation
  • Distributed Debate
  • Distributed Parallel Optimization
  • 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 --pre

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.

Refer to our tutorial for more details.

Tutorial

• About AgentScope
• Installation
• Quick Start
• Model
• Prompt Engineering
• Agent
• Memory
• Response Parser
• Tool
• Pipeline and MsgHub
• Distribution
• AgentScope Studio
• Logging
• Monitor
• Example: Werewolf Game

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
             Xuchen Pan and
             Weirui Kuang and
             Zhijian Ma and
             Bingchen Qian and
             Fei Wei and
             Wenhao Zhang and
             Yuexiang Xie and
             Daoyuan Chen and
             Liuyi Yao and
             Hongyi Peng and
             Ze Yu Zhang 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},
}