langfengQ/verl-agent

Agent Frameworks

verl-agent is an extension of veRL, designed for training LLM/VLM agents via RL. verl-agent is also the official code for paper "Group-in-Group Policy Optimization for LLM Agent Training"

GitHub →Website →

6.9

GitHub Metrics

Stars

2.1k

Forks

200

Open Issues

Watchers

Contributors

108

Weekly Commits

Language

Python

License

Apache-2.0

Last Commit

Jun 9, 2026

Created

Mar 23, 2025

Latest Release

v0.1.0

Release Date

Dec 11, 2025

Synced: Jun 29, 2026

Quality Scores

Documentation Qualityw: 20%

7.4

Has docs site (https://huggingface.co/papers/2505.10978). Description: 187 chars. Stars signal: 2,053. Contributors: 108. Score: 7.4/10

Community Healthw: 20%

6.0

Stars: 2,053. Contributors: 108. Watchers: 7. Forks: 200. Issue ratio: 3.2%. Score: 6/10

Maintenance Velocityw: 15%

6.3

Last commit: 19d ago. Weekly commits: 0. Latest release: v0.1.0. Maturity bonus: 1.3y old. Score: 6.3/10

API Design & DXw: 20%

7.3

Stars/issues ratio: 31. Dynamic language: Python. Has documentation site. Permissive license: Apache-2.0. Popularity signal: 2,053 stars. Score: 7.3/10

Production Readinessw: 15%

7.1

Battle-tested: 2,053 stars. Peer review: 108 contributors. Versioned: v0.1.0. Licensed: Apache-2.0. Age: 1.3 years. Maintenance: last commit 19d ago. Score: 7.1/10

Ecosystem Integrationw: 10%

7.4

Fork interest: 200. Major ecosystem: Python. Integration-friendly: Apache-2.0. Adoption: 2,053 stars. Has web presence. Score: 7.4/10