Skip to main content

OpenRLHF

An Easy-to-use, Scalable and High-performance Agentic RL Framework based on Ray (PPO & DAPO & REINFORCE++ & VLM & TIS & vLLM & Ray & Async RL)

View on GitHub Official site
Fine Tuning Training Apache-2.0 Medium setup 9,833 stars

Overview

Description

Build better AI assistants that follow instructions, reason effectively, and align with human preferences. OpenRLHF is a production-ready open-source framework that scales reinforcement learning training from human feedback to hundreds of billions of parameters. The bottom line: used by Google, ByteDance, and other leading AI companies to train state-of-the-art language models.

Technical scorecard

License

Apache-2.0

Commercial use

Yes

OpenAI-compatible API

Yes

REST API

Yes

Fine-tuning support

Yes

Quantization support

Yes

Docker available

Yes

GUI / no-code available

No

Telemetry

Unknown

Offline after setup

Yes

Data & Privacy

Does it send data online?

After setup, this listing is marked as usable offline. Confirm network behavior against the upstream project before regulated deployment.

Does it store history?

Not verified in this directory yet. Review the upstream docs for persistence, logs, and workspace storage.

License checks?

Commercial use is marked as allowed or likely allowed by the listed license.

Telemetry?

Unknown

Last verified: Jul 22, 2026. Maintainer verification should be treated as directory guidance, not legal advice.

🛡️

Deploying OpenRLHF in production?

Stop silent telemetry leaks before they trigger an unintended data leak. Get our Airgap Certainty Blueprint for OpenRLHF.

Request Privacy and Telemetry Audit

Setup & Installation

Medium

A developer can usually get this running with standard docs.

# Start with the official project repository
# https://github.com/OpenRLHF/OpenRLHF

Hardware Requirements

Hardware tagsGPU (NVIDIA)
LanguagesPython

Works Well With