菜单
NVIDIA CUTLASS CUDA Template Library - comprehensive reference for high-performance matrix multiplication (GEMM), convolution, tensor operations, and CuTe DSL across all GPU architectures (Volta through Blackwell)
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
基于 SOC 职业分类
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/jstzwj/ai-infra-plugins --skill cutlass命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
正在显示 SKILL.md
Comprehensive reference documentation and skill for xFormers, Facebook Research's toolbox to accelerate research on Transformers. Use this skill whenever the user mentions xformers, memory_efficient_attention, FMHA, flash attention, SwiGLU, RMSNorm, RoPE, rope_padded, 2:4 structured sparsity, sparsify24, sequence parallelism, fused all-gather/reduce-scatter, tiled matmul, block-sparse tensors, attention patterns, selective activation checkpointing, forward-backward overlap, tree attention, model parallel linear layers, xformers profiler, Triton kernels for transformers, CUTLASS attention, BlockDiagonalMask, LowerTriangularMask, merge_attentions, or xformers internals and build configuration.
Comprehensive reference documentation and skill for bitsandbytes, the k-bit quantization library for PyTorch enabling accessible large language models. Use this skill whenever the user mentions bitsandbytes, LLM.int8(), QLoRA, 4-bit quantization, 8-bit quantization, NF4, FP4, 8-bit optimizers, block-wise quantization, Int8Params, Params4bit, Linear8bitLt, Linear4bit, StableEmbedding, quantize_blockwise, dequantize_blockwise, quantize_4bit, dequantize_4bit, quantize_nf4, quantize_fp4, GlobalOptimManager, paged optimizers, FSDP integration with quantization, Triton kernels for quantization, CPU/XPU/MPS/HPU backends, CUDA kernels for quantized matmul, or bitsandbytes internals.
NVIDIA Megatron-LM & Megatron Core - GPU-optimized framework for training large language models with tensor parallelism, pipeline parallelism, data parallelism (DDP/FSDP), context parallelism, expert parallelism, FP8/FP4 quantization, CUDA graphs, MoE (Mixture of Experts), multimodal models, and TensorRT-LLM export. Supports GPT, BERT, T5, Mamba, LLaMA, Mixtral, DeepSeek-V3, Qwen3, and custom architectures from 2B to 462B parameters with up to 47% MFU on H100 GPUs.
| name | cutlass |
| description | NVIDIA CUTLASS CUDA Template Library - comprehensive reference for high-performance matrix multiplication (GEMM), convolution, tensor operations, and CuTe DSL across all GPU architectures (Volta through Blackwell) |
| version | 3.8 |
CUTLASS (CUDA Templates for Linear Algebra Subroutines) is NVIDIA's open-source C++ template library for implementing high-performance matrix-matrix multiplication (GEMM) and related computations at all levels and scales within CUDA. It provides near-optimal utilization of peak theoretical throughput on NVIDIA GPUs.
CUTLASS provides:
Use this skill when:
#include "cutlass/cutlass.h"
#include "cutlass/gemm/device/gemm_universal_adapter.h"
#include "cutlass/gemm/kernel/gemm_universal.h"
#include "cutlass/epilogue/collective/default_epilogue.hpp"
#include "cutlass/gemm/collective/collective_builder.hpp"
// Define types
using ElementA = cutlass::half_t;
using ElementB = cutlass::half_t;
using ElementC = float;
using ElementD = cutlass::half_t;
using LayoutA = cutlass::layout::RowMajor;
using LayoutB = cutlass::layout::ColumnMajor;
using LayoutC = cutlass::layout::RowMajor;
using LayoutD = cutlass::layout::RowMajor;
// Use CollectiveBuilder for convenient kernel assembly
using CollectiveOp = typename cutlass::gemm::collective::CollectiveBuilder<
cutlass::arch::Sm90, cutlass::arch::OpClassTensorOp,
ElementA, LayoutA, 8,
ElementB, LayoutB, 8,
ElementC,
cutlass::gemm::GemmShape<128, 128, 64>,
cutlass::gemm::collective::StageCountAutoCarveout<0>,
cutlass::gemm::collective::KernelScheduleAuto
>::CollectiveOp;
using EpilogueOp = cutlass::epilogue::collective::DefaultEpilogue<
LayoutD, LayoutC,
cutlass::epilogue::collective::EpilogueScheduleAuto
>;
using Kernel = cutlass::gemm::kernel::GemmUniversal<
cutlass::gemm::collective::CollectiveOp,
cutlass::epilogue::collective::EpilogueOp
>;
using Gemm = cutlass::gemm::device::GemmUniversalAdapter<Kernel>;
Gemm gemm_op;
args = Gemm::Arguments{
{M, N, K},
{ptr_A, stride_A},
{ptr_B, stride_B},
{ptr_C, stride_C},
{ptr_D, stride_D},
{alpha, beta}
};
(args);
#include "cutlass/gemm/device/gemm.h"
using Gemm = cutlass::gemm::device::Gemm<
cutlass::half_t, cutlass::layout::RowMajor, // ElementA, LayoutA
cutlass::half_t, cutlass::layout::ColumnMajor, // ElementB, LayoutB
float, cutlass::layout::RowMajor, // ElementC, LayoutC
float, // ElementAccumulator
cutlass::arch::OpClassTensorOp, // OpClass
cutlass::arch::Sm80, // ArchTag
cutlass::gemm::GemmShape<128, 128, 32>, // ThreadblockShape
cutlass::gemm::GemmShape<64, 64, 32>, // WarpShape
cutlass::gemm::GemmShape<16, 8, 16>, // InstructionShape
cutlass::epilogue::thread::LinearCombination<
float, 4, float, float>
>;
Gemm gemm_op;
gemm_op({M, N, K}, alpha, ptr_A, lda, ptr_B, ldb, beta, ptr_C, ldc, ptr_D, ldd);
| Architecture | SM Version | Key Features |
|---|---|---|
| Volta | SM70 | First Tensor Cores (FP16), WMMA API |
| Turing | SM75 | Tensor Cores (INT8/INT4/INT1), MMA instructions |
| Ampere | SM80/SM89 | TF32, BF16, async copy, cp.async |
| Hopper | SM90 | TMA, GMMA, WGMMA, thread block clusters, warp specialization |
| Blackwell | SM100+ | UMMA, block-scaled types (NVFP4/MXFP), distributed GEMM, green contexts |
Detailed reference documentation is organized in the references/ directory: