Openrlhf Training
技能 已验证 活跃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.
To enable efficient and high-performance Reinforcement Learning from Human Feedback (RLHF) training for large language models using a distributed architecture with advanced acceleration techniques.
功能
- High-performance RLHF training framework
- Support for PPO, GRPO, RLOO, DPO algorithms
- Ray + vLLM acceleration for large models (7B-70B+)
- Distributed architecture with multi-node GPU cluster support
- Hybrid Engine for GPU resource sharing
使用场景
- Training large language models with RLHF
- Fine-tuning models on custom reward functions
- Leveraging distributed computing for faster training
- Accelerating inference during RLHF rollout phases
非目标
- Single-node or basic model fine-tuning
- Environments without GPU acceleration capabilities
- Inference-only model serving outside of the training loop
安装
npx skills add davila7/claude-code-templates通过 npx 运行 Vercel skills CLI(skills.sh)— 需要本地安装 Node.js,以及至少一个兼容 skills 的智能体(Claude Code、Cursor、Codex 等)。前提是仓库遵循 agentskills.io 格式。
质量评分
已验证类似扩展
Openrlhf Training
99High-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.
Verl Rl Training
99Provides guidance for training LLMs with reinforcement learning using verl (Volcano Engine RL). Use when implementing RLHF, GRPO, PPO, or other RL algorithms for LLM post-training at scale with flexible infrastructure backends.
Moe Training
98Train Mixture of Experts (MoE) models using DeepSpeed or HuggingFace. Use when training large-scale models with limited compute (5× cost reduction vs dense models), implementing sparse architectures like Mixtral 8x7B or DeepSeek-V3, or scaling model capacity without proportional compute increase. Covers MoE architectures, routing mechanisms, load balancing, expert parallelism, and inference optimization.
Ray Data
95Scalable data processing for ML workloads. Streaming execution across CPU/GPU, supports Parquet/CSV/JSON/images. Integrates with Ray Train, PyTorch, TensorFlow. Scales from single machine to 100s of nodes. Use for batch inference, data preprocessing, multi-modal data loading, or distributed ETL pipelines.
Verl Rl Training
95Provides guidance for training LLMs with reinforcement learning using verl (Volcano Engine RL). Use when implementing RLHF, GRPO, PPO, or other RL algorithms for LLM post-training at scale with flexible infrastructure backends.
PyTorch Lightning
100Deep learning framework (PyTorch Lightning). Organize PyTorch code into LightningModules, configure Trainers for multi-GPU/TPU, implement data pipelines, callbacks, logging (W&B, TensorBoard), distributed training (DDP, FSDP, DeepSpeed), for scalable neural network training.