Practical guidance for training MoE VLMs in Megatron Bridge. Compares FSDP and 3D-parallel approaches, using rounded lessons from Qwen3-VL, Qwen3-Next, and other multimodal experiments.
Install
npx skills add NVIDIA/skills --skill nemo-mbridge-perf-moe-vlm-trainingnpx skills add NVIDIA/skills --skill nemo-mbridge-perf-moe-vlm-training -a claude-code -gnpx skills add NVIDIA/skills --skill nemo-mbridge-perf-moe-vlm-training -a codex -ggemini skills install https://github.com/NVIDIA/skills.git --path skills/nemo-mbridge-perf-moe-vlm-traininggh skill install NVIDIA/skills nemo-mbridge-perf-moe-vlm-traininggit 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-moe-vlm-training ~/.claude/skills/nemo-mbridge-perf-moe-vlm-training # Claude Code
cp -R /tmp/NVIDIA__skills/skills/nemo-mbridge-perf-moe-vlm-training ~/.agents/skills/nemo-mbridge-perf-moe-vlm-training # Codex, Cursor, Gemini CLI, Copilot, OpenCode