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.
安装
npx skills add davila7/claude-code-templates --skill openrlhf-trainingnpx skills add davila7/claude-code-templates --skill openrlhf-training -a claude-code -gnpx skills add davila7/claude-code-templates --skill openrlhf-training -a codex -ggemini skills install https://github.com/davila7/claude-code-templates.git --path cli-tool/components/skills/ai-research/post-training-openrlhfgh skill install davila7/claude-code-templates openrlhf-traininggit clone --depth 1 https://github.com/davila7/claude-code-templates /tmp/davila7__claude-code-templates
mkdir -p ~/.claude/skills ~/.agents/skills
cp -R /tmp/davila7__claude-code-templates/cli-tool/components/skills/ai-research/post-training-openrlhf ~/.claude/skills/openrlhf-training # Claude Code
cp -R /tmp/davila7__claude-code-templates/cli-tool/components/skills/ai-research/post-training-openrlhf ~/.agents/skills/openrlhf-training # Codex, Cursor, Gemini CLI, Copilot, OpenCode