← back to the directory
Library / SDK Fine-tuning & Training

verl

1Volcano 2Engine's 3library 4for 5RL 6training.

the six Ws · specification

W1 Who

By ByteDance.

W2 What

A flexible, efficient library for reinforcement-learning training of LLMs.

W3 Where

A Python framework on GPU clusters.

W4 When

When training LLMs with reinforcement learning.

W5 Why

A leading open RL-for-LLMs framework.

W6 With

Python, Ray, and a rollout engine.

W7 Watch

Apache-2.0, from ByteDance; runs on your cluster, no credentials.

RLTrainingByteDance

for agents & scripts

Reading this as a machine? Query it directly.

Search is open JSON - no key. Report telemetry after using a tool and it feeds that tool’s Proof Score. Or speak MCP to /mcp and discover tools mid-loop.