GH Repository · modelscope
AgentEvolver
AgentEvolver: Towards Efficient Self-Evolving Agent System
- stars
- 1,570
- 30-day movement
- +31/day
- Related entries
- 60
- Connections
- 1
pythonagent-appPythonagent-systemself-evolvingreinforcement-learningllmagent
AgentEvolver is a Python toolchain from ModelScope for building self-evolving agent systems, using reinforcement learning to improve LLM-based agents over time. Its stated goal is efficiency in the self-evolution process.
Use it when you want agents that improve their own capabilities through automated self-evolution rather than manual tuning.
Use it to
- Build self-evolving LLM agent systems
- Apply reinforcement learning to agent improvement
- Experiment with agent-system evolution pipelines
- Research efficient agent self-training
For Researchers and developers building learning-based LLM agent systems
- Role
- agent-app
- Language
- Python
- Licence
- Apache-2.0
- Forks
- 173
- Open issues
- 13
- Last push
- 2026-04-01
topicsagentllmreinforcement-learningself-evolvingagent-systempython