EvoAgentX/Awesome-Self-Evolving-Agents
A survey paper categorizing approaches to self-evolving AI agents across single-agent, multi-agent, and domain-specific optimization from 2023 to 2025, accompanied by the EvoAgentX open-source framework.
Not currently ranked — collecting fresh signals.
star history
This repository provides a visual taxonomy and conceptual framework for understanding how AI agents evolve and optimize themselves over time. It categorizes research into single-agent optimization, multi-agent optimization, and domain-specific optimization approaches. The repository includes a companion open-source framework called EvoAgentX for automated evolution of agentic workflows, along with references to related research papers and methods in the space.
Frequently asked
- What is EvoAgentX/Awesome-Self-Evolving-Agents?
- A survey paper categorizing approaches to self-evolving AI agents across single-agent, multi-agent, and domain-specific optimization from 2023 to 2025, accompanied by the EvoAgentX open-source framework.
- Is Awesome-Self-Evolving-Agents open source?
- Yes — EvoAgentX/Awesome-Self-Evolving-Agents is open source, released under the MIT license.
- How popular is Awesome-Self-Evolving-Agents?
- EvoAgentX/Awesome-Self-Evolving-Agents has 2.4k stars on GitHub.
- Where can I find Awesome-Self-Evolving-Agents?
- EvoAgentX/Awesome-Self-Evolving-Agents is on GitHub at https://github.com/EvoAgentX/Awesome-Self-Evolving-Agents.