EvoAgentX - Building a Self-Evolving Ecosystem of AI Agents

May 15, 2025

Go to Project Site

Building a Self-Evolving Ecosystem of AI Agents

English | 简体中文

An automated framework for evaluating and evolving agentic workflows.

🔥 Latest News

[May 2025] 🎉 EvoAgentX has been officially released!

⚡ Get Started

Installation
LLM Configuration
Automatic WorkFlow Generation
Demo Video
Evolution Algorithms
Applications
Tutorial and Use Cases

Installation

We recommend installing EvoAgentX using pip:

pip install git+https://github.com/EvoAgentX/EvoAgentX.git

For local development or detailed setup (e.g., using conda), refer to the Installation Guide for EvoAgentX.

Example (optional, for local development):

git clone https://github.com/EvoAgentX/EvoAgentX.git
cd EvoAgentX
# Create a new conda environment
conda create -n evoagentx python=3.10

# Activate the environment
conda activate evoagentx

# Install the package
pip install -r requirements.txt
# OR install in development mode
pip install -e .

LLM Configuration

API Key Configuration

To use LLMs with EvoAgentX (e.g., OpenAI), you must set up your API key.

Option 1: Set API Key via Environment Variable

Linux/macOS:

export OPENAI_API_KEY=<your-openai-api-key>

Windows Command Prompt:

set OPENAI_API_KEY=<your-openai-api-key>

Windows PowerShell:

$env:OPENAI_API_KEY="<your-openai-api-key>" # " is required

Once set, you can access the key in your Python code with:

import os
OPENAI_API_KEY = os.getenv("OPENAI_API_KEY")

Option 2: Use .env File

Create a .env file in your project root and add the following:

OPENAI_API_KEY=<your-openai-api-key>

Then load it in Python:

from dotenv import load_dotenv 
import os 

load_dotenv() # Loads environment variables from .env file
OPENAI_API_KEY = os.getenv("OPENAI_API_KEY")

Configure and Use the LLM

Once the API key is set, initialise the LLM with:

from evoagentx.models import OpenAILLMConfig, OpenAILLM

# Load the API key from environment
OPENAI_API_KEY = os.getenv("OPENAI_API_KEY")

# Define LLM configuration
openai_config = OpenAILLMConfig(
    model="gpt-4o-mini",       # Specify the model name
    openai_key=OPENAI_API_KEY, # Pass the key directly
    stream=True,               # Enable streaming response
    output_response=True       # Print response to stdout
)

# Initialize the language model
llm = OpenAILLM(config=openai_config)

# Generate a response from the LLM
response = llm.generate(prompt="What is Agentic Workflow?")

📖 More details on supported models and config options: LLM module guide.

Automatic WorkFlow Generation

Once your API key and language model are configured, you can automatically generate and execute multi-agent workflows in EvoAgentX.

🧩 Core Steps:

Define a natural language goal
Generate the workflow with WorkFlowGenerator
Instantiate agents using AgentManager
Execute the workflow via WorkFlow

💡 Minimal Example:

from evoagentx.workflow import WorkFlowGenerator, WorkFlowGraph, WorkFlow
from evoagentx.agents import AgentManager

goal = "Generate html code for the Tetris game"
workflow_graph = WorkFlowGenerator(llm=llm).generate_workflow(goal)

agent_manager = AgentManager()
agent_manager.add_agents_from_workflow(workflow_graph, llm_config=openai_config)

workflow = WorkFlow(graph=workflow_graph, agent_manager=agent_manager, llm=llm)
output = workflow.execute()
print(output)

You can also:

📊 Visualise the workflow: workflow_graph.display()
💾 Save/load workflows: save_module() / from_file()

📂 For a complete working example, check out the workflow_demo.py

Demo Video

🎥 Demo video:

In this demo, we showcase the workflow generation and execution capabilities of EvoAgentX through two examples:

Application 1: Intelligent Job Recommendation from Resume
Application 2: Visual Analysis of A-Share Stocks

✨ Final Results

Application 1:
Job Recommendation

Application 2:
Stock Visual Analysis

Evolution Algorithms

We have integrated some existing agent/workflow evolution algorithms into EvoAgentX, including TextGrad, MIPRO and AFlow.

To evaluate the performance, we use them to optimize the same agent system on three different tasks: multi-hop QA (HotPotQA), code generation (MBPP) and reasoning (MATH). We randomly sample 50 examples for validation and other 100 examples for testing.

Tip: We have integrated these benchmark and evaluation code in EvoAgentX. Please refer to the benchmark and evaluation tutorial for more details.

📊 Results

Method	HotPotQA (F1%)	MBPP (Pass@1 %)	MATH (Solve Rate %)
Original	63.58	69.00	66.00
TextGrad	69.21	68.00	76.00
AFlow	65.09	79.00	71.00
MIPRO	69.16	68.00	72.30

Please refer to the examples/optimization folder for more details.

Applications

We use our framework to optimize existing multi-agent systems on the GAIA benchmark. We select Open Deep Research and OWL, two representative multi-agent framework from the GAIA leaderboard that is open-source and runnable.

We apply EvoAgentX to optimize their prompts. The performance of the optimized agents on the GAIA benchmark validation set is shown in the figure below.

Open Deep Research

OWL Agent

Full Optimization Reports: Open Deep Research and OWL.

Tutorial and Use Cases

💡 New to EvoAgentX? Start with the Quickstart Guide for a step-by-step introduction.

Explore how to effectively use EvoAgentX with the following resources:

Cookbook	Description
Build Your First Agent	Quickly create and manage agents with multi-action capabilities.
Build Your First Workflow	Learn to build collaborative workflows with multiple agents.
Automatic Workflow Generation	Automatically generate workflows from natural language goals.
Benchmark and Evaluation Tutorial	Evaluate agent performance using benchmark datasets.
AFlow Optimizer Tutorial	Automatically optimise multi-agent workflows for better task performance.

🛠️ Follow the tutorials to build and optimize your EvoAgentX workflows.

🚀 We’re actively working on expanding our library of use cases and optimization strategies. More coming soon — stay tuned!

🎯 Roadmap

Modularize Evolution Algorithms: Abstract optimization algorithms into plug-and-play modules that can be easily integrated into custom workflows.
Develop Task Templates and Agent Modules: Build reusable templates for typical tasks and standardized agent components to streamline application development.
Integrate Self-Evolving Agent Algorithms: Incorporate more recent and advanced agent self-evolution across multiple dimensions, including prompt tuning, workflow structures, and memory modules.
Enable Visual Workflow Editing Interface: Provide a visual interface for workflow structure display and editing to improve usability and debugging.

🙋 Support

Join the Community

📢 Stay connected and be part of the EvoAgentX journey!
🚩 Join our community to get the latest updates, share your ideas, and collaborate with AI enthusiasts worldwide.

Discord — Chat, discuss, and collaborate in real-time.
X (formerly Twitter) — Follow us for news, updates, and insights.
WeChat — Connect with our Chinese community.

Contact Information

If you have any questions or feedback about this project, please feel free to contact us. We highly appreciate your suggestions!

Email: evoagentx.ai@gmail.com

We will respond to all questions within 2-3 business days.

🙌 Contributing to EvoAgentX

Thanks go to these awesome contributors

We appreciate your interest in contributing to our open-source initiative. We provide a document of contributing guidelines which outlines the steps for contributing to EvoAgentX. Please refer to this guide to ensure smooth collaboration and successful contributions. 🤝🚀

📚 Acknowledgements

This project builds upon several outstanding open-source projects: AFlow, TextGrad, DSPy, LiveCodeBench, and more. We would like to thank the developers and maintainers of these frameworks for their valuable contributions to the open-source community.

📄 License

Source code in this repository is made available under the MIT License.

AI Agent