Download
AgentEvolver is an open-source research framework for building self-evolving AI agents powered by large language models. The system focuses on improving the efficiency and scalability of training autonomous agents by allowing them to generate tasks, explore environments, and refine strategies without heavy reliance on manually curated datasets. Its architecture combines reinforcement learning with LLM-driven reasoning mechanisms to guide exploration and learning. The framework introduces several key mechanisms, including self-questioning to create new learning tasks, self-navigating to improve exploration through experience reuse, and self-attributing to assign rewards based on the usefulness of actions. These mechanisms enable agents to continuously improve their capabilities while interacting with complex environments and tools. AgentEvolver also integrates environment sandboxes, experience management systems, and modular data pipelines to support large-scale experimentation.
Features
- Self-evolving agent framework combining reinforcement learning and LLM reasoning
- Environment sandboxing for interactive agent training experiments
- Exploration optimization using self-navigation and experience reuse
- Reward attribution system for improving sample efficiency
- Curiosity-driven task generation through self-questioning mechanisms
- Modular architecture for integrating tools, models, and datasets
Project Samples
