Training & fine-tuning
verl
Coordinates training and rollout engines through a hybrid execution model with flexible resource mapping.
Overview
Research summary
verl implements the HybridFlow approach to large-language-model reinforcement learning. It separates computational work and data dependencies so that training backends and generation engines can be composed through modular interfaces. Documented integrations include FSDP, Megatron-LM, vLLM and SGLang, with algorithms such as PPO and GRPO. Its execution model supports flexible placement of components across GPUs and transitions between training and generation phases. The project is maintained by the verl community after originating within ByteDance’s Seed team.
Repository summary
- Stars
- 23,724
- Open issues
- Unavailable
- Last push
- 2026-10-01
- Commits, 90 days
- Unavailable
- Repository activity
- Not scored
- Version
- Unavailable
Recorded catalogue figures. View repository data and provenance →
Classification
Pricing & services
Paid services unknown
Whether the provider offers paid products or services has not been established.
Licence scope
Implementation
Recorded implementation details and interfaces for verl.
Implementation details
Python library and supporting tools
- Languages
- Python
- Repository type
- source
Recorded interfaces and capabilities
Licence scope
Repository
Repository snapshots, release information and recorded maintenance signals.
Repository snapshot
- Stars
- 23,724
- Open issues
- Unavailable
- Last push
- 2026-10-01
- Commits, 90 days
- Unavailable
- Repository activity
- Not scored
- Archived
- Not recorded
Repository activity is a snapshot, not a quality or popularity ranking. It combines recent-push freshness (50%), 90-day commits (30%) and issue pressure (20%).
Maintenance and provenance
- Catalogue snapshot
- 2026-10-06
- Stars source
- Recorded fallback
- Last push source
- Recorded fallback
Documentation
Recorded references and research provenance for this entry.
Recorded sources 6
- https://verl-project.github.io Project page
- https://github.com/verl-project/verl Linked repository · Research reference
- https://verl.readthedocs.io/en/latest/ Documentation
- https://api.github.com/repos/verl-project/verl Research reference
- https://github.com/verl-project/verl/blob/main/README.md Research reference
- https://github.com/verl-project/verl/blob/main/LICENSE Research reference
Research metadata
- Research date
- 2026-10-02
- Catalogue snapshot
- 2026-10-06
Community & social 1 channels
Communities
- GitHub Discussionsgithub.com/verl-project/verl/discussionsOfficial
Link details for GitHub Discussions
- Last checked