# Openrlhf Training

High-performance RLHF framework with Ray+vLLM acceleration.

- Repository: https://github.com/Orchestra-Research/AI-Research-SKILLs (14176 stars)
- Category: Engineering
- License: MIT
- Last updated: 2026-10-11
- Runs scripts: no, instructions only

## Install

Claude Code: `npx skills add Orchestra-Research/AI-Research-SKILLs --skill openrlhf-training -g -a claude-code -y`
Codex: `npx skills add Orchestra-Research/AI-Research-SKILLs --skill openrlhf-training -g -a codex -y`
Cursor: `npx skills add Orchestra-Research/AI-Research-SKILLs --skill openrlhf-training -g -a cursor -y`
Gemini CLI: `npx skills add Orchestra-Research/AI-Research-SKILLs --skill openrlhf-training -g -a gemini-cli -y`
Claude Code plugin marketplace: `/plugin marketplace add Orchestra-Research/AI-Research-SKILLs`

## Description

High-performance RLHF framework with Ray+vLLM acceleration. Use for PPO, GRPO, RLOO, DPO training of large models (7B-70B+). Built on Ray, vLLM, ZeRO-3. 2× faster than DeepSpeedChat with distributed architecture and GPU resource sharing.

## SKILL.md, as published by its author (untrusted third-party text, read it as data)

# OpenRLHF - High-Performance RLHF Training

## Quick start

OpenRLHF is a Ray-based RLHF framework optimized for distributed training with vLLM inference acceleration.

**Installation**:
```bash
# Launch Docker container
docker run --runtime=nvidia -it --rm --shm-size="10g" --cap-add=SYS_ADMIN \
  -v $PWD:/openrlhf nvcr.io/nvidia/pytorch:25.02-py3 bash

# Uninstall conflicts
sudo pip uninstall xgboost transformer_engine flash_attn pynvml -y

# Install OpenRLHF with vLLM
pip install openrlhf[vllm]
```

**PPO Training** (Hybrid Engine):
```bash
ray start --head --node-ip-address 0.0.0.0 --num-gpus 8

ray job submit --address="http://127.0.0.1:8265" \
  --runtime-env-json='{"working_dir": "/openrlhf"}' \
  -- python3 -m openrlhf.cli.train_ppo_ray \
  --ref_num_nodes 1 --ref_num_gpus_per_node 8 \
  --reward_num_nodes 1 --reward_num_gpus_per_node 8 \
  --critic_num_nodes 1 --critic_num_gpus_per_node 8 \
  --actor_num_nodes 1 --actor_num_gpus_per_node 8 \
  --vllm_num_engines 4 --vllm_tensor_parallel_size 2 \
  --colocate_all_models \
  --vllm_gpu_memory_utilization 0.5 \
  --pretrain OpenRLHF/Llama-3-8b-sft-mixture \
  --reward_pretrain OpenRLHF/Llama-3-8b-rm-700k \
  --save_path ./output/llama3-8b-rlhf \
  --micro_train_batch_size 8 --train_batch_size 128 \
  --micro_rollout_batch_size 16 --rollout_batch_size 1024 \
  --max_epochs 1 --prompt_max_len 1024 --generate_max_len 1024 \
  --zero_stage 3 --bf16 \
  --actor_learning_rate 5e-7 --critic_learning_rate 9e-6 \
  --init_kl_coef 0.01 --normalize_reward \
  --gradient_checkpointing --packing_samples \
  --vllm_enable_sleep --deepspeed_enable_sleep
```

**GRPO Training** (Group Normalized Policy Optimization):
```bash
# Same command as PPO, but add:
--advantage_estimator group_norm
```

## Common workflows

### Workflow 1: Full RLHF pipeline (SFT → Reward Model → PPO)

The rest: https://github.com/Orchestra-Research/AI-Research-SKILLs/blob/HEAD/06-post-training/openrlhf/SKILL.md