Dev & Engineering ✓ NVIDIA · Official tilegymhugging-face-transformerscuda-kernelsmonkey-patchingllm-inferencellm-training

TileGym Transformers Kernel Integration

Integrate TileGym kernels into Hugging Face Transformers models without editing Transformers source.

FollowSkills review · FSRS-2.0
Not recommended
51/ 100 5-point scale 2.6 / 5
Trust16 / 25 · 3.2/5

The documentation requires user confirmation before experimentation, constrains GPU selection by UUID, mandates Docker, limits editable paths, and requires baseline comparison, correctness checks, and rollback attempts. However, least-privilege scope, data-flow disclosure, recovery boundaries, and the implications of branch/reset, GPU execution, web lookup, and autonomous code generation remain incomplete, so 9 points are deducted.

Reliability7 / 20 · 1.8/5

The workflow specifies environment constraints, timeouts, baselines, metrics, and failure branches, including abandoning invalid candidates. However, no reproducible implementation tests are included for this skill path, dependency availability and diagnostic failure messages are only partly covered, and static calibration caps the score at 10; 13 points are deducted.

Adaptability9 / 15 · 3.0/5

The audience, trigger semantics, primary input (model ID), and GPU/Docker environment are fairly clear, with supported GPU architectures stated. Non-fit model ranges, no-GPU alternatives, Chinese interaction, and mainland-China network reachability are not adequately addressed, so 6 points are deducted.

Convention10 / 15 · 3.3/5

The material is layered into workflow, environment, integration, auto-kernelization, and inventory-schema references, and includes version, author, license, dependencies, and risks. The recommended Examples section is missing, while changelog, maintenance ownership, and update path are unclear; 5 points are deducted.

Effectiveness6 / 15 · 2.0/5

The core goal, experiment loop, coverage/throughput constraint, and results-record format are defined and could guide integration. However, there is no directly usable implementation example for this skill, model adaptation is costly, and performance/correctness evidence is reported rather than statically reproducible, so 9 points are deducted.

Verifiability3 / 10 · 1.5/5

The skill defines citation requirements and repository path constraints and includes a benchmark report and evaluation tasks. The cited material was not verified within the supplied evidence, and there is no committed test suite or CI covering key paths; benchmark claims do not provide independent reproduction, so 7 points are deducted.

Evidence confidence:Low Reviewed Jul 20, 2026 Reviewed revision 55f18499943e
The upstream repository has new commits since this review. The score still applies to the reviewed revision shown and may not cover the latest changes.
Before you use it
  • The skill guides branch operations, Docker/GPU experiments, autonomous code generation, and commits; confirm the worktree is clean and define acceptable changes, resource use, and rollback scope before execution.
  • Performance, coverage, and correctness results are presented as report material; no experiments were executed during this static review, so independently validate them on the target model, CUDA/cuTile versions, and GPU.
  • The workflow depends on an NVIDIA GPU, Docker, CUDA tooling, Transformers/Hugging Face, and potentially overseas documentation or model services; no mainland-China fallback is specified.
See the full review method →

What does this skill do, and when should you use it?

This skill guides an AI agent in integrating TileGym kernels into Hugging Face transformers models. It uses non-intrusive monkey patching to replace selected modules, classes, and initialization, forward, and weight-loading methods without directly modifying the transformers source. The workflow also creates new cuTile kernels for uncovered PyTorch code and evaluates end-to-end correctness and performance. The skill is specifically intended for TileGym and transformers integration, and is stated to be verified with specified Claude Code, Codex, and Cursor model environments.

Guides the agent through experiment-environment preparation, existing-kernel integration, and autonomous kernelization; replaces selected transformers modules, classes, and methods for a target Hugging Face model; applies patches before model instantiation so runtime execution uses TileGym implementations; creates new cuTile kernels for uncovered PyTorch code; records reusable transformer-local kernels using FlashInfer-Bench-style Definition and Solution metadata; and summarizes the resulting work.

  1. An engineer validating TileGym kernel correctness in a specified Hugging Face language model.
  2. A developer who needs to replace model components without modifying the transformers source tree.
  3. A performance engineer measuring end-to-end throughput effects of TileGym kernels during LLM training or inference.
  4. A researcher creating cuTile kernels for PyTorch operations not yet covered in a target model.

What are this skill's strengths and limitations?

Pros
  • Uses a non-intrusive monkey-patch approach instead of directly editing transformers source.
  • Covers both integration of existing TileGym kernels and creation of new cuTile kernels.
  • Addresses end-to-end functional correctness, kernel coverage, and throughput improvement.
  • Requires structured Definition and Solution metadata for reusable kernels.
Limitations
  • The supplied material does not identify supported models or transformers versions.
  • The detailed environment, integration, and auto-kernelization procedures are in references files that were not provided.
  • No concrete performance figures or test results are included.
  • Monkey patching depends on the target model's internal transformers modules and methods, so implementation changes may affect integration.

How do you install this skill?

Install the skill from NVIDIA/skills with the skills CLI:

npx skills add nvidia/skills --skill tilegym-monkey-patch-kernels-to-transformers --yes

You can target Codex explicitly:

npx skills add nvidia/skills --skill tilegym-monkey-patch-kernels-to-transformers --agent codex

The supplied source does not document a manual-copy destination.

How do you use this skill?

Give a loaded agent the skill name and target model ID, for example:

Hi, please /monkey-patch-kernels-to-transformers Qwen/Qwen3.5-0.8B.

The agent may ask clarification questions before proceeding. The supplied SKILL.md does not include the detailed environment commands, supported-model list, or failure-recovery procedures.

How does this skill compare with similar options?

The source identifies direct modification of the transformers source as an alternative; this skill instead uses non-intrusive monkey patching before model instantiation.

FAQ

Does it modify the transformers source code?
No. Its described approach replaces selected modules, classes, and methods before model instantiation.
Can I use any Hugging Face model?
The material gives a target model ID as an example but does not promise universal model support; feasibility depends on the target implementation and available integrable kernels.
Do I need to clone the repository to install it?
The README's default flow uses npx skills add, so cloning NVIDIA/skills is not required.
Does it guarantee a performance improvement?
No guarantee is stated. The workflow evaluates correctness and performance, but the supplied material provides no quantified uplift.

More skills from this repository

All from NVIDIA/skills

Dev & Engineering ✓ NVIDIA · Official

TileGym cuTile Kernel Optimizer

Systematically reduce cuTile GPU kernel latency through profiling, bottleneck diagnosis, and controlled tuning.

Dev & Engineering ✓ NVIDIA · Official

TileGym cuTile Kernel Guide

A step-by-step guide for registering, implementing, testing, and benchmarking a new cuTile GPU kernel in TileGym.

Automation & Ops ✓ NVIDIA · Official

Jetson LLM Serving

Deploy OpenAI-compatible LLM and VLM servers on Jetson with vLLM or SGLang.

Dev & Engineering ✓ NVIDIA · Official

NV-Reason-CXR Chest X-ray Reasoning Test

Runs reproducible command-shape and live inference smoke tests for chest X-ray reasoning.

Dev & Engineering ✓ NVIDIA · Official

NeMo MBridge GPU Memory Tuning

Diagnose GPU OOMs in Megatron Bridge training and reduce fragmentation, activation, and PEFT memory usage.

Dev & Engineering ✓ NVIDIA · Official

NeMo AutoModel Model Onboarding

A structured guide for adding new LLM, MoE, and VLM architectures to NeMo AutoModel.

Dev & Engineering ✓ NVIDIA · Official

Megatron Bridge Activation Recompute

Trade extra compute for lower GPU activation memory in Megatron Bridge training.

Dev & Engineering ✓ NVIDIA · Official

NeMo MBridge MoE VLM Training Guide

Guides FSDP and 3D-parallel choices and tuning for MoE vision-language model training.

Dev & Engineering ✓ NVIDIA · Official

NeMo MBridge CUDA Graphs

Configure, validate, and benchmark CUDA Graph capture in Megatron Bridge to reduce host-driver overhead.

Dev & Engineering ✓ NVIDIA · Official

Earth2Studio Deterministic Forecast

Build single-member weather forecast inference scripts with Earth2Studio.

Dev & Engineering ✓ NVIDIA · Official

NeMo MBridge MoE Performance Workflow

Systematically diagnose and improve MoE training throughput in Megatron Bridge.

Dev & Engineering ✓ NVIDIA · Official

CuTile Autotuning Guide

Design, implement, and validate low-overhead autotuning for CuTile GPU kernels.

Dev & Engineering ✓ NVIDIA · Official

NeMo MBridge Hierarchical Context Parallelism

Enable, troubleshoot, and verify hierarchical context parallelism in Megatron-Bridge.

Dev & Engineering ✓ NVIDIA · Official

Jetson Inference Memory Tuning

Choose a Jetson serving runtime and generate memory flags from the device snapshot and workload.

Automation & Ops ✓ NVIDIA · Official

NeMo-RL Brev Runbook

Helps NeMo-RL agents manage storage, caches, logs, and authentication safely on Brev instances.

Dev & Engineering ✓ NVIDIA · Official

NeMo MBridge Long-Context MoE Training

Practical Megatron Bridge guidance for sizing parallelism, managing memory, and preserving throughput in long-context MoE training.

Data & Analysis ✓ NVIDIA · Official

Synthetic Brain MRI Generator

Generate synthetic brain MRI volumes through NVIDIA’s documented workflow.

Dev & Engineering ✓ NVIDIA · Official

cuTile Python-to-Julia Converter

Port cuTile Python GPU kernels into tested cuTile.jl Julia implementations.

Dev & Engineering ✓ NVIDIA · Official

cuTile Python Kernel Assistant

Guides developers in writing, debugging, validating, and optimizing high-performance cuTile GPU kernels.

Dev & Engineering ✓ NVIDIA · Official

cuTile-to-Triton Kernel Converter

Port cuTile GPU kernels to validated, tunable Triton implementations.

Related skills