From: verl-project <github.com/verl-project>Newsgroups: alt.freeweights.toolsSubject:

verl

Release: v0.9.1 (2026-09-20)Commit: 1876b06License: Apache-2.0

Reinforcement learning library for post-training language models at scale.

trainingv0.9.1releaseverlPythonApache-2.0
FactsLicenseApache-2.0LanguagePythonKindtrainingRelease2026-09-20PlatformsLinux (GPU)Links to the projectpermissive
$ pip install verl

Overview

Reinforcement learning library for post-training language models at scale.

Kind
training
License
Apache-2.0
Language
Python
Platforms
Linux (GPU)
Latest release
v0.9.1 (2026-09-20)
Repository
verl-project/verl

Freeweights links to the project and does not host tool binaries.

Install

$ pip install verl
$ pip install verl

Checked against the project's README on 2026-11-14. The project's own instructions win when they differ: install docs

Freeweights links to the project and does not host tool binaries.

Use with Freeweights

Fine-tune a catalog model

Download the weights from Freeweights first (HF_ENDPOINT or fw.py), then point the tool at the folder.

Catalog models it can fine-tune (3137)

and 3077 more.

License

License
Apache-2.0
Class
Permissive

Read the license

Freeweights links to the project and does not host tool binaries.

Provenance

From: verl-project <github.com/verl-project>Newsgroups: alt.freeweights.toolsSubject: verlRelease: v0.9.1 (2026-09-20)Commit: 1876b06License: Apache-2.0
Repository
verl-project/verl
Pinned release
v0.9.1, 2026-09-20
Release commit
1876b06d0a3e4e71e06230be10af14492ca8a75b
Default branch
main 4191e7f910d85d69acb8c45f4840f10faa535165
Fetched
2026-11-14
Takedowns
None.

Freeweights links to the project and does not host tool binaries.

Works with

3137 catalog models. See the list