openrlhf-training
✓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.
Installation
SKILL.md
OpenRLHF is a Ray-based RLHF framework optimized for distributed training with vLLM inference acceleration.
Workflow 1: Full RLHF pipeline (SFT → Reward Model → PPO)
Hybrid Engine GPU sharing: See references/hybrid-engine.md for vLLM sleep mode, DeepSpeed sleep mode, and optimal node allocation.
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. Source: orchestra-research/ai-research-skills.
Facts (cite-ready)
Stable fields and commands for AI/search citations.
- Install command
npx skills add https://github.com/orchestra-research/ai-research-skills --skill openrlhf-training- Category
- </>Dev Tools
- Verified
- ✓
- First Seen
- 2026-02-11
- Updated
- 2026-02-18
Quick answers
What is openrlhf-training?
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. Source: orchestra-research/ai-research-skills.
How do I install openrlhf-training?
Open your terminal or command line tool (Terminal, iTerm, Windows Terminal, etc.) Copy and run this command: npx skills add https://github.com/orchestra-research/ai-research-skills --skill openrlhf-training Once installed, the skill will be automatically configured in your AI coding environment and ready to use in Claude Code or Cursor
Where is the source repository?
https://github.com/orchestra-research/ai-research-skills
Details
- Category
- </>Dev Tools
- Source
- skills.sh
- First Seen
- 2026-02-11