Loading...
Loading...
Found 418 Skills
Writing tests for GPUI applications. Use when testing components, async operations, or UI behavior.
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.
Julia: multiple dispatch, type system, metaprogramming, Pkg, scientific computing, GPU CUDA.jl
Vast.ai CLI to manage GPU instances, volumes, serverless endpoints, and billing.
Validate and use selective and full activation recompute in Megatron Bridge to reduce GPU memory usage at the cost of extra compute.
Use this skill when building, deploying, evaluating, debugging, or measuring latency for the DeepStream SOP Inference Microservice — a GPU-accelerated FastAPI service that detects whether operators perform assembly-line steps in order via event boundary detection (GEBD) plus VLM classification. Trigger even if the user does not name it: verify operator step sequence, detect missing or out-of-order SOP steps, score factory/work-cell video for procedure compliance, run VLM-based SOP checking on industrial cameras, or call /v1/chat/completions with a file, RTSP, or Basler camera. Also trigger for its internals: SOPVideoProcessor, DeepStream GEBD model (e.g. DDM) via Triton CAPI, nvds_custom_postprocess, Cosmos Reason 1/2 vLLM, SSE streaming, Kafka NvProto/JSON output, Basler/Pylon camera + emulation, Docker compose, chunk-level latency. Do NOT trigger for generic DeepStream pipelines, object detection/tracking, NIM imports, or video summarization.
GPU-accelerated data curation for LLM training. Supports text/image/video/audio. Features fuzzy deduplication (16× faster), quality filtering (30+ heuristics), semantic deduplication, PII redaction, NSFW detection. Scales across GPUs with RAPIDS. Use for preparing high-quality training datasets, cleaning web data, or deduplicating large corpora.
Create serverless endpoint templates and endpoints on RunPod.io. Supports Python/Node.js runtimes, GPU selection (3090, A100, etc.), and idempotent configuration. Use this skill when a user wants to set up a new serverless endpoint or template on RunPod.
Generates Rust code for GPUI desktop UI components following Zed editor patterns. Use when building desktop applications with gpui crate, creating themed UI components, implementing autocomplete/completions, building command palettes, or working with the gpui-component library. Covers RenderOnce components, Entity state management, theming with ActiveTheme, and Zed-style UI patterns.
GPUI UI framework best practices for building desktop applications. Use when writing GPUI code, creating UI components, handling state/events, async tasks, animations, lists, forms, testing, or working with Zed-style Rust GUI code.
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.
Add a new cuTile GPU kernel operator to TileGym. Covers dispatch registration in ops.py, cuTile backend implementation, __init__.py exports, test creation, and benchmark in tests/benchmark. Use when adding, creating, or implementing a new cuTile operator/kernel in TileGym, or when asking how to register a new cuTile op.