Nemo Mbridge Perf Cuda Graphs
nvidia/skills/skills/nemo-mbridge-perf-cuda-graphsby nvidiacf5224d14250Apache-2.03.5K starsListed Oct 8, 2026Updated Oct 8, 2026Repository updated today
Validate and use CUDA graph capture in Megatron Bridge, including local full-iteration graphs and Transformer Engine scoped graphs for attention, MLP, and MoE modules.
- cf5224d14250Currentcommit cf5224dPublished Oct 8, 2026
Source and attribution
Source:nvidia/skillsinskills/nemo-mbridge-perf-cuda-graphsat commitcf5224d
License: Apache-2.0
Content belongs to its original authors. SourceWeft indexes it from a public repository.
More from nvidia/skills
Tilegym Cutile Python
nvidia
Guides writing, debugging and optimizing cuTile Python GPU kernels, with examples, validation loops and optional multi-agent orchestration.
Tilegym Monkey Patch Kernels To Transformers
nvidia
Integrates TileGym cuTile kernels into Hugging Face transformers models via non-intrusive monkey-patching and auto-generates new kernels.
Tilegym Converting Cutile To Julia
nvidia
Converts cuTile Python GPU kernels to cuTile.jl Julia equivalents, with rules, examples and a static validator.
Tilegym Adding Cutile Kernel
nvidia
Guides adding a new cuTile GPU kernel operator to TileGym, from dispatch registration through tests and benchmarks.
Physicsnemo Discover
nvidia
Guides discovery of PhysicsNeMo model families, datapipes and examples for a SciML task, citing live repo paths.
Nemoclaw User Guide
nvidia
Routes an AI agent to NemoClaw's canonical Markdown docs and docs MCP server to answer setup and operations questions.
More in Software Development
Playground
anthropics
Builds self-contained interactive HTML playgrounds with controls, live preview, and copyable prompt output.
M5 Onboard
anthropics
Provisions M5Stack ESP32 boards by detecting them on USB, flashing UIFlow 2.0 firmware, and installing a MicroPython app bundle.
Cardputer Buddy
anthropics
Guides iterating on a Cardputer-Adv MicroPython app bundle: adding apps, pushing files over USB-serial, tailing logs, and running REPL commands.
Web Design Guidelines
vercel-labs
Reviews UI code against Web Interface Guidelines fetched from a remote source and reports findings as file:line entries.
Vercel React Native Skills
vercel-labs
React Native and Expo best-practice rules covering list performance, animation, navigation, UI patterns, state, and monorepo setup.
Vercel React Best Practices
vercel-labs
Vercel's React and Next.js performance guidelines, organized as 70 rules for writing, reviewing, and refactoring code.