68 skills · sorted by relevance
Deploy language models with vLLM through an OpenAI-compatible API or offline batch processing. Tune response speed, throughput, and GPU memory use with quantization, which reduces numerical precision, and execution across multiple GPUs.
Deploy language model APIs and run batch generation with vLLM. Configure GPU memory use, quantization—lower precision model storage—and distribution across GPUs, then test response times and monitor how many requests the server handles.
Serves LLMs with high throughput using vLLM's PagedAttention and continuous batching. Use when deploying production LLM APIs, optimizing inference latency/throughput, or serving models with limited GPU memory. Supports OpenAI-compatible endpoints, quantization (GPTQ/AWQ/FP8), and tensor parallelism.
The English summary is being prepared. Showing the original description.
Pick the right ComfyUI startup flags for VRAM, attention, caching, and speed. The full decision matrix for OOM (--novram / --cache-none / --disable-smart-memory), shared-VRAM creep on Windows (--reserve-vram N), model-switching with big text encoders (--cache-none), high-VRAM throughput (--gpu-only / --highvram), and attention-backend selection (--use-sage-attention for speed, --use-pytorch-cross-attention as the highest-quality / Z-Image-safe fallback). Also the acceleration-stack + Blackwell/RTX 5000 (sm_120) notes. Use when a graph OOMs (especially long video like LTX 2 / WAN), when the GPU spills into shared VRAM and slows to a crawl, when switching between models eats all RAM, when Z-Image produces black/garbled output under Sage, or when deciding which attention backend to launch with. Flag names verified against upstream comfy/cli_args.py; see Sources.
The English summary is being prepared. Showing the original description.
Use when you need to refactor Java code for high performance — including memory/allocation reduction, CPU hot-path optimization, and syntax/API/control-flow improvements. This should trigger for requests such as Review Java code for high performance; Optimize Java hot path; Reduce Java allocations; Improve Java latency/throughput. Part of Plinth Toolkit
The English summary is being prepared. Showing the original description.
Expert-level materials informatics covering materials databases, machine learning for property prediction, high-throughput computation, and data-driven materials discovery.
The English summary is being prepared. Showing the original description.
Serve LLMs com alta throughput usando PagedAttention do vLLM e continuous batching. Use ao fazer deploy de APIs LLM em produção, otimizar latência/throughput de inferência, ou servir modelos com memória GPU limitada. Suporta endpoints compatíveis com OpenAI, quantização (GPTQ/AWQ/FP8) e tensor parallelism.
The English summary is being prepared. Showing the original description.
Set up vLLM to serve large language models through an OpenAI-compatible API or process prompts in batches. The skill covers deployment, performance monitoring, and quantization, which reduces model memory requirements.
Configure LLM inference and serving on NVIDIA GPUs with TensorRT-LLM. Covers quantization, which reduces numerical precision, batching requests during generation, and distributing models across GPUs to improve processing capacity and response times.
Configure LLM inference and serving on NVIDIA GPUs with TensorRT-LLM. Covers quantization, which reduces numerical precision, batching requests during generation, and distributing models across GPUs to improve processing capacity and response times.
The sva package contains functions for removing batch effects and other unwanted variation in high-throughput experiment. Specifically, the sva package contains functions for the identifying and building surrogate variables for high-dimensi
The English summary is being prepared. Showing the original description.
Systematically find and fix performance problems in applications, frontend and backend, by measuring first. Use this skill whenever the user reports slow endpoints, high latency, poor throughput, memory leaks, large bundles, or slow page loads; wants to profile with cProfile, py-spy, Chrome DevTools, Node --cpu-prof, or perf; run load tests with locust or k6; set performance budgets or SLOs; optimize database queries (EXPLAIN, indexes, N+1); add caching (Redis); improve frontend performance (code splitting, lazy loading, image optimization, memoization, virtualization); or tune the network (CDNs, gzip/br, HTTP/2/3). Also trigger for p50/p95/p99 latency, throughput, error rate, connection pooling, async vs threads, GIL, and memory leak investigations.
The English summary is being prepared. Showing the original description.
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.
The English summary is being prepared. Showing the original description.
Serves LLMs with high throughput using vLLM's PagedAttention and continuous batching. Use when deploying production LLM APIs, optimizing inference latency/throughput, or serving models with limited GPU memory. Supports OpenAI-compatible endpoints, quantization (GPTQ/AWQ/FP8), and tensor parallelism.
The English summary is being prepared. Showing the original description.
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.
The English summary is being prepared. Showing the original description.
Set up and optimize large language model inference, or response generation, on NVIDIA GPUs. Use quantization, batch processing, and multiple GPUs to improve serving performance, and expose models through a chat completion API.
Deploy Suricata to detect suspicious network traffic or block it inline. Configure protocol inspection, detection rules, and structured EVE JSON logs, then connect logs to analysis platforms and check packet drops and alert quality during operation.
Assess Google Cloud workloads for environmental sustainability using the Well-Architected Framework. Get recommendations on lower-carbon regions, efficient computing and storage, AI energy use, and ongoing emissions measurement with Carbon Footprint.
Use this skill when the user is measuring the host or DPU-CPU control-plane rate of a DOCA Flow pipeline with doca_flow_perf — picking a JSON policy from configs/, choosing the DPDK or DOCA backend, running the single-iteration smoke then the iterative eval loop, interpreting per-iteration CPU cycles and num_pushed / num_failed, or capturing the four-tuple (DOCA version, BlueField/firmware, JSON policy, worker/queue/burst config) that makes a Kops/sec number defensible. Trigger even when the user does not explicitly mention "doca-flow-perf" — typical implicit phrasings include "how many rules per second can my BlueField insert", "5-tuple hairpin rule rate", "Kops/sec for steering", "flow-perf number does not match release notes", "DPDK vs DOCA benchmark", or "rule-install variance too high". Refuse and route elsewhere for optimizing a live Flow app (doca-flow-tune), the DPA-offloaded path (doca-flow-dpa-perf), dataplane throughput or latency, or library-internal pipe semantics — those belong to other skills.
The English summary is being prepared. Showing the original description.
[omh] Accepted plan awaiting implementation: split it into disjoint parallel lanes with per-lane acceptance criteria, verification commands, and owners; prevents two lanes editing the same file. Aliases: ulw. Use when the user says: ultrawork, parallel work, parallel implementation, parallel then integrate, high throughput, coding team, coordinated workers, finish until done.
The English summary is being prepared. Showing the original description.
Build event streaming applications using Azure Event Hubs SDK for JavaScript (@azure/event-hubs). Use when implementing high-throughput event ingestion, real-time analytics, IoT telemetry, or event-driven architectures with partitioned consumers.
The English summary is being prepared. Showing the original description.
Build real-time streaming applications with Azure Event Hubs SDK for Java. Use when implementing event streaming, high-throughput data ingestion, or building event-driven architectures.
The English summary is being prepared. Showing the original description.
Azure Event Hubs SDK for Python streaming. Use for high-throughput event ingestion, producers, consumers, and checkpointing. Triggers: "event hubs", "EventHubProducerClient", "EventHubConsumerClient", "streaming", "partitions".
The English summary is being prepared. Showing the original description.
Azure Event Hubs SDK for .NET. Use for high-throughput event streaming: sending events (EventHubProducerClient, EventHubBufferedProducerClient), receiving events (EventProcessorClient with checkpointing), partition management, and real-time data ingestion. Triggers: "Event Hubs", "event streaming", "EventHubProducerClient", "EventProcessorClient", "send events", "receive events", "checkpointing", "partition".
The English summary is being prepared. Showing the original description.