datamllab /
HarborRL
An end-to-end framework for Agentic Reinforcement Learning.
Updated 7 min ago
+83 stars gained this week
Stars gained each day, last 30 days
Oct 9Oct 24
This week Before
In short
It is aimed at researchers training tool-using agents on sandboxed tasks, and it expects Python with a GPU for real training. The README covers a CPU demo via harborrl hello, a configuration wizard, and a local Web console.
Summary written by AI from the repository’s description and README.
datamllab / HarborRL
View on GitHub(opens in new tab)