Validate and use selective and full activation recompute in Megatron Bridge to reduce GPU memory usage at the cost of extra compute. Use for activation memory OOMs or regressions involving recompute_granularity, recompute_num_layers, recompute_modules, recompute_method, selective recompute, full recompute, or activation checkpointing.
Install
npx skills add NVIDIA/skills --skill nemo-mbridge-perf-activation-recomputenpx skills add NVIDIA/skills --skill nemo-mbridge-perf-activation-recompute -a claude-code -gnpx skills add NVIDIA/skills --skill nemo-mbridge-perf-activation-recompute -a codex -ggemini skills install https://github.com/NVIDIA/skills.git --path skills/nemo-mbridge-perf-activation-recomputegh skill install NVIDIA/skills nemo-mbridge-perf-activation-recomputegit clone --depth 1 https://github.com/NVIDIA/skills /tmp/NVIDIA__skills
mkdir -p ~/.claude/skills ~/.agents/skills
cp -R /tmp/NVIDIA__skills/skills/nemo-mbridge-perf-activation-recompute ~/.claude/skills/nemo-mbridge-perf-activation-recompute # Claude Code
cp -R /tmp/NVIDIA__skills/skills/nemo-mbridge-perf-activation-recompute ~/.agents/skills/nemo-mbridge-perf-activation-recompute # Codex, Cursor, Gemini CLI, Copilot, OpenCode