Every premium and members-only resource for $9.99/month, as a no-commitment subscription. See the plans

PalateKit

Sign in
Menu

Request a resource

Describe the resource you are missing in one sentence. We will email you when it is published.

0 / 280

A page, a video or an image online.

Unlock PalateKit

Unlock everything

Plan

Paying means accepting our terms of sale.

Secure payment by Stripe. Receipts and invoices sent by email.

Welcome back

Sign in to install, download or save this resource. A free account is enough for free resources.

Training models with distributed RLHF using OpenRLHF v1.0.0

Use this resource to train large models with PPO, GRPO, RLOO or DPO across several machines

New

  • 478 downloads
  • 206 likes
  • 216 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install openrlhf-distributed-rlhf-training

With no option, the skill goes to ~/.agents/skills/; add --agent claude-code (or codex, cursor…) to target an agent.

First time: connect the CLI with

npx --yes palatekit@latest login

The agent guides you through starting the Docker container, installing OpenRLHF with vLLM, then chaining an SFT, reward model and PPO pipeline, or running GRPO or DPO training. You get launch commands and common workflows.

Mostly picked by

Updated November 2, 2026

Card for the “Training models with distributed RLHF using OpenRLHF” skill

Details

Works in
Install guides
Generates with
Your agent's model
Licence
MIT

Similar resources

View all

What you get

  • SKILL.md: installation, PPO, GRPO and DPO workflows
  • references/multi-node-training.md, custom-rewards.md, hybrid-engine.md and algorithm-comparison.md: multi-node, rewards, hybrid engine and comparison

Good fit for

  • You want a full RLHF pipeline across several GPUs
  • You want GRPO training without a critic model

Not for

    Tags

    • rlhf
    • ppo
    • grpo
    • distributed

    About

    You say

    “Set up a GRPO run with OpenRLHF on 8 GPUs for my reasoning model.”

    Card for the “Training models with distributed RLHF using OpenRLHF” skill

    Skill published by Orchestra Research under the MIT license, delivered in its original version.

    Show more

    How to use it

    1. Install the skill in your agent's skills folder
    2. Ask your agent to train large models with PPO, GRPO, RLOO or DPO across several machines
    3. Review and adjust the result

    Content preview

    Files

    • LICENSE1.1 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • references/algorithm-comparison.md9.9 KB
    • references/custom-rewards.md16 KB
    • references/hybrid-engine.md7.4 KB
    • references/multi-node-training.md11.3 KB
    • SKILL.md8.4 KB

    Origin and licence

    Listed unmodified

    Author
    Orchestra Research
    Source
    github.com/orchestra-research/ai-research-skills/tree/773a52944ba4747a18bd4ae9ade53fff041adcbc/06-post-training/openrlhf
    Licence
    MIT

    Notice

    « openrlhf-training » (Orchestra Research, MIT) livré dans sa version d'origine : fiche en français et en anglais par PalateKit.

    User reviews

    –

    No ratings yet

    Loading reviews…

    Delete your review?

    Your rating and comment will be removed from the page.