Loading...
Loading...
Found 9 Skills
Provides guidance for writing and benchmarking optimized CUDA kernels for NVIDIA GPUs (H100, A100, T4) targeting HuggingFace diffusers and transformers libraries. Supports models like LTX-Video, Stable Diffusion, LLaMA, Mistral, and Qwen. Includes integration with HuggingFace Kernels Hub (get_kernel) for loading pre-compiled kernels. Includes benchmarking scripts to compare kernel performance against baseline implementations.
Guides Holoscan SDK installation: inspects the host, assesses platform compatibility, recommends an install method, and delegates to the matching install skill.
Trains large language models (2B-462B parameters) using NVIDIA Megatron-Core with advanced parallelism strategies. Use when training models >1B parameters, need maximum GPU efficiency (47% MFU on H100), or require tensor/pipeline/sequence/context/expert parallelism. Production-ready framework used for Nemotron, LLaMA, DeepSeek.
Fine-tune any HuggingFace CV / VLM / LLM model on local NVIDIA GPUs inside an NGC PyTorch container. Use when the user wants to fine-tune a HuggingFace model (full or LoRA), train a vision / VLM / LLM model end-to-end, generate a reproducible HF training pipeline, smoke-test a HuggingFace model locally before scale-up, push a fine-tuned model to the HF Hub with a model card, or emit a self-contained rerun skill for an existing HuggingFace finetune. Supports image classification, object detection, semantic / instance / panoptic segmentation, depth estimation, image-text-to-text VLM (SFT / LoRA), and LLM SFT / DPO / GRPO. Six-step workflow: inspect and qualify, hardware and NGC image, research, generate and smoke, train + eval + infer, push and emit rerun skill.
Deploy vLLM using Docker (pre-built images or build-from-source) with NVIDIA GPU support and run the OpenAI-compatible server.
Local Docker execution for TAO SDK job containers using the host Docker daemon and NVIDIA GPU runtime. Use when running TAO jobs on the current machine or a directly attached Docker host. Trigger phrases include "run locally", "local Docker", "use my GPU", "run on my machine", "host Docker daemon".
Build Holoscan SDK from source via the in-tree ./run script. Use only when published packages don't meet the user's needs.
Optimizes LLM inference with NVIDIA TensorRT for maximum throughput and lowest latency. Use for production deployment on NVIDIA GPUs (A100/H100), when you need 10-100x faster inference than PyTorch, or for serving models with quantization (FP8/INT4), in-flight batching, and multi-GPU scaling.
CuTe Python DSL API reference and implementation patterns for NVIDIA GPU kernel programming. Provides execution model, core API table, key constraints, common patterns, and documentation index. Use when: (1) writing or modifying CuTe DSL kernel code, (2) looking up CuTe DSL API syntax, (3) implementing attention/GEMM/MLA patterns in CuTe DSL, (4) understanding CuTe DSL execution model and compilation pipeline, (5) checking what CuTe DSL can and cannot do.