GPU and accelerator computing

Covers the rules for writing Gradio Spaces on ZeroGPU hardware: the @spaces.GPU decorator, duration and quota tuning, process isolation and CUDA build limits.

huggingface/skills

Adds verified layer guides such as L0 and L1 and compact GPU lanes to an existing Torch Profiler Chrome trace, changing how it looks but not how it ran.

BBuf/AI-Infra-Auto-Driven-SKILLS

A skill your agent uses for GPU training or inference clusters on SageMaker HyperPod (Slurm or EKS), ParallelCluster, or self-managed EC2/EKS GPU instances.

aws/tools-for-devops-agent

Diagnose performance issues on Amazon SageMaker HyperPod clusters — uneven NCCL bandwidth across nodes and poor filesystem throughput.

awslabs/agent-plugins

Write, debug, and optimize Triton and Gluon GPU kernels from local upstream tutorials, production kernels, language definitions, and compiler source.

slowlyC/agent-gpu-skills

Analyze single-file PyTorch/Kineto Chrome trace .json(.gz) files using the VeloQ CLI.

lucifer1004/VeloQ

Draft, debug, and measure CUDA kernels and host launch workflows.

sablin39/tilelang-cuda-skills

Implements request-to-token-pool copies for Triton kernels on Ascend NPUs, replacing loop-carried vector offsets with fixed-offset block loops that compile reliably.

Krusty84/triton-ascend-agent-dev-kit

Patterns and pitfalls for the ONNX-domain Attention operator's CUDA implementation in ONNX Runtime: dispatch cascade, eligibility limits, mask and bias kernels, and test routing.

microsoft/onnxruntime

Convert PyTorch ATDISPATCH macros to ATDISPATCHV2 format in ATen C++ code.

intel/torch-xpu-ops

Loads large language models in 8-bit or 4-bit with bitsandbytes so they fit smaller GPUs, and sets up QLoRA fine-tuning on a 4-bit base model.

Orchestra-Research/AI-Research-SKILLs

Scales PyTorch, TensorFlow and Hugging Face training from a single GPU to multi-node clusters with Ray Train, including Ray Tune sweeps and checkpoint recovery.

Orchestra-Research/AI-Research-SKILLs

Sets up large-scale LLM training with NVIDIA Megatron-Core, choosing tensor, pipeline, data, context and expert parallelism for a given model size and GPU count.

Orchestra-Research/AI-Research-SKILLs

Detects CPU, GPU, memory and disk resources before heavy scientific tasks and writes a JSON file with advice on parallelism, out-of-core work and GPU use.

davila7/claude-code-templates

Organizes PyTorch training code into LightningModules, DataModules and Trainers, with multi-GPU strategies, callbacks and logging configured.

davila7/claude-code-templates

Review 3DGS implementation code for correctness, performance bugs, and best practices.

jaccen/Awesome-Gaussian-Skills

Writes a Triton kernel pattern for Ascend NPU that gathers several indexed token rows per program into an on-chip buffer and stores one contiguous output tile, for MoE-style token reordering.

Krusty84/triton-ascend-agent-dev-kit

Breaks LLM torch profiler traces down by forward pass, layer and kernel, with timing tables and Perfetto time ranges for the layers you want to inspect.

BBuf/AI-Infra-Auto-Driven-SKILLS

Guide to using Mamba selective state-space models for linear-time sequence modeling, from the Mamba block and pretrained checkpoints to Mamba-2 and speed comparisons.

Orchestra-Research/AI-Research-SKILLs

Orchestrates video data augmentation and auto-labeling workflows on OSMO, from flow selection and preflight checks to submission, monitoring and output download.

NVIDIA/skills

PyTorch training reference: architecture choice by data type, scaling rules, a training loop, optimizer and learning-rate choices, and fixes for loss spikes or OOM.

Orchestra-Research/AI-Research-SKILLs

Guide to renting GPUs on Lambda Labs for ML training and inference: on-demand instances, 1-Click Clusters, SSH access, persistent filesystems and alternatives.

Orchestra-Research/AI-Research-SKILLs

Write optimized Triton GPU kernels for deep learning operations.

vipshop/cache-dit

Compares SGLang, vLLM, TensorRT-LLM and TokenSpeed on one model and workload, searching server flags to find the best deployment command within a latency SLA.

BBuf/AI-Infra-Auto-Driven-SKILLS

Recover from GPU-busy / GPU-unavailable failures. An agent skill from facebookexperimental/triton.

facebookexperimental/triton

Builds Triton-Ascend kernels that pack sorted token assignments into fixed expert-capacity tensors, restore top-k outputs and compute router-weight gradients on Ascend NPUs.

Krusty84/triton-ascend-agent-dev-kit

Guided step-by-step wizard for producing a new FlyDSL GPU kernel from a requirement: classify the kernel type, pick a skeleton, fill in compute, add control flow / sync / LDS, then test on GPU.

ROCm/FlyDSL

Benchmarks LLM inference and drives GPU kernel optimization with Magpie.

amd/skills

Debug Triton compilation by dumping IR at each stage (TTIR, TTGIR, LLVM, PTX).

facebookexperimental/triton

Guides building a FlashAttention-v2-style fused forward attention kernel for Triton-Ascend on Ascend NPU, with online softmax, causal staging and float32 accumulation.

Krusty84/triton-ascend-agent-dev-kit

Sets up and runs NVIDIA Cosmos Policy evaluations on the LIBERO and RoboCasa simulators, including headless GPU rendering and inference latency profiling.

Orchestra-Research/AI-Research-SKILLs

Fine-tunes and evaluates OpenVLA-OFT and OFT+ robot policies with LoRA and continuous action heads on LIBERO simulation and ALOHA real-robot setups.

Orchestra-Research/AI-Research-SKILLs

Fine-tunes and serves Physical Intelligence's pi0, pi0-fast and pi0.5 robot policies with JAX or PyTorch, including checkpoint conversion and policy servers.

Orchestra-Research/AI-Research-SKILLs

Step-by-step tutorial for adding a new lightweight JIT CUDA kernel to sglang.kernels JIT infrastructure and public operator groups

sgl-project/sglang

Estimate peak GPU memory for a DualPipeV training run. An agent skill from mlc-ai/pith-train.

mlc-ai/pith-train

3DGS compression-to-deployment pipeline: quantization (scalar/VQ/mixed-precision), pruning (coreset/adaptive/variational/merge/Bayesian), progressive streaming & LoD, Web/WebGPU/mobile deployment…

jaccen/Awesome-Gaussian-Skills

Builds a fused row-wise softmax kernel for Triton-Ascend that reads and writes each row once, handling padding, strides and masked loads on Ascend NPUs.

Krusty84/triton-ascend-agent-dev-kit

Optimize dense matrix multiplication (GEMM) kernels in Triton for NVIDIA and AMD GPUs.

ZJLi2013/awesome-kernel-skills

Query live GPU inventory, submit an authenticated Itô fixed-rate RFQ, inspect RFQ or procurement status, revoke device credentials, and run explicitly gated node qualification through the separately…

affaan-m/ECC

Check and compare software component versions on SageMaker HyperPod cluster nodes - NVIDIA drivers, CUDA toolkit, cuDNN, NCCL, EFA, AWS OFI NCCL, GDRCopy, MPI, Neuron SDK (Trainium/Inferentia)…

awslabs/agent-plugins

Write, optimize, and debug high-performance AI compute kernels using TileLang (a Python DSL for GPU programming).

yzlnew/infra-skills

Modal is a serverless cloud platform for running Python on demand, including on-demand GPUs.

K-Dense-AI/scientific-agent-skills

GPU-accelerates scientific Python on NVIDIA hardware and verifies that the result is correct and faster.

K-Dense-AI/scientific-agent-skills

Operate GPU-backed Kubernetes clusters for AI inference and training with scheduling, autoscaling, node health, MIG partitioning, and cost controls.

sickn33/agentic-awesome-skills

A skill your agent uses when the user asks "why is this B200/Blackwell kernel slow, what should I optimize first, should I fuse or move to Tensor Cores/TMA".

mirage-project/mirage

A skill your agent uses when a B200/Blackwell kernel shows wrong results, uncoalesced global memory access, SMEM bank conflicts, a TMA swizzle that mismatches the Tensor Core read, or confused…

mirage-project/mirage

A skill your agent uses when you need a NEW or optimized MPK CUDA kernel (a per-task .cuh under include/mirage/persistentkernel/tasks/) that must PROVABLY beat a target WITHOUT the kernel-optimizer…

mirage-project/mirage

Diagnose NCCL failures and adjacent training-pod failures on HyperPod GPU clusters (EKS or Slurm) — training hangs, AllReduce / collective-op timeouts, EFA or libfabric errors, rendezvous failures…

awslabs/agent-plugins