179 件 ・ 関連度順
概要と使いどころ
Activation-aware weight quantization for 4-bit LLM compression with 3x speedup and minimal accuracy loss. Use when deploying large models (7B-70B) on limited GPU memory, when you need faster inference than GPTQ with better accuracy preservation, or for instruction-tuned and multimodal models. MLSys 2024 Best Paper Award winner.
日本語の概要は準備中です。原文の説明を表示しています。
Activation-aware weight quantization for 4-bit LLM compression with 3x speedup and minimal accuracy loss. Use when deploying large models (7B-70B) on limited GPU memory, when you need faster inference than GPTQ with better accuracy preservation, or for instruction-tuned and multimodal models. MLSys 2024 Best Paper Award winner.
日本語の概要は準備中です。原文の説明を表示しています。
Expert guidance for fast fine-tuning with Unsloth - 2-5x faster training, 50-80% less memory, LoRA/QLoRA optimization
日本語の概要は準備中です。原文の説明を表示しています。
Post-training 4-bit quantization for LLMs with minimal accuracy loss. Use for deploying large models (70B, 405B) on consumer GPUs, when you need 4× memory reduction with <2% perplexity degradation, or for faster inference (3-4× speedup) vs FP16. Integrates with transformers and PEFT for QLoRA fine-tuning.
日本語の概要は準備中です。原文の説明を表示しています。
Provides React Native performance optimization guidelines for FPS, TTI, bundle size, memory leaks, re-renders, and animations. Applies to tasks involving Hermes optimization, JS thread blocking, bridge overhead, FlashList, native modules, or debugging jank and frame drops.
日本語の概要は準備中です。原文の説明を表示しています。
Expert guidance for fast fine-tuning with Unsloth - 2-5x faster training, 50-80% less memory, LoRA/QLoRA optimization
日本語の概要は準備中です。原文の説明を表示しています。
Golang performance optimization patterns and methodology - if X bottleneck, then apply Y. Covers allocation reduction, CPU efficiency, memory layout, GC tuning, pooling, caching, and hot-path optimization. Use when profiling or benchmarks have identified a bottleneck and you need the right optimization pattern to fix it. Also use when performing performance code review to suggest improvements or benchmarks that could help identify quick performance gains. Not for measurement methodology (→ See `samber/cc-skills-golang@golang-benchmark` skill) or debugging workflow (→ See `samber/cc-skills-golang@golang-troubleshooting` skill).
日本語の概要は準備中です。原文の説明を表示しています。
Expert guidance for fast fine-tuning with Unsloth - 2-5x faster training, 50-80% less memory, LoRA/QLoRA optimization
日本語の概要は準備中です。原文の説明を表示しています。
Profiles Python code for performance bottlenecks and memory issues. Use when Python code is slow or when profiling for optimization before a release.
日本語の概要は準備中です。原文の説明を表示しています。
Performance profiling and optimization for web apps — Core Web Vitals (LCP, INP, CLS), Lighthouse audits, bundle analysis, backend profiling (CPU, memory, DB queries), N+1 detection, caching strategies (Redis, CDN, HTTP), and performance budgets. Use when user asks to improve performance, run Lighthouse audit, profile a Node.js app, optimize Core Web Vitals, reduce bundle size, or investigate slow response times. Do NOT use for database schema optimization (use db-sculptor), Docker image optimization (use docker), or CDN configuration.
日本語の概要は準備中です。原文の説明を表示しています。
Audits and scores Claude Projects using a six-dimension Scorecard with anchored 1-5 rubrics and detects seven structural anti-patterns. When global layer info is provided, adds cross-layer alignment findings. Use when user says "audit my project," "review my custom instructions," "score my project," "evaluate my Claude project," "improve my system prompt," "what's wrong with my project," "why is my project underperforming." Also trigger on symptom-phrased: "my project doesn't work right," "Claude is inconsistent in my project," "my Project used to work and now doesn't." Also use when the user pastes Custom Instructions asking why output is poor or generic. Do NOT use when the user's primary request is a global-layer audit (use rootnode-global-audit if available), single-prompt evaluation (use rootnode-prompt-validation if available), or Memory-only optimization (use rootnode-memory-optimization if available). Run on Opus 5 or Sonnet 5 at `high` effort (both defaults); depth reduces on legacy models.
日本語の概要は準備中です。原文の説明を表示しています。
Spark, pandas, polars, DuckDB optimization for batch data processing. Activate on: batch processing, Spark optimization, polars, DuckDB, pandas performance, data frame, shuffle, partition, memory optimization. NOT for: streaming pipelines (use streaming-pipeline-architect), warehouse queries (use data-warehouse-optimizer).
日本語の概要は準備中です。原文の説明を表示しています。
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.
日本語の概要は準備中です。原文の説明を表示しています。
Otimiza atenção em transformers com Flash Attention para ganho de 2-4x em velocidade e redução de 10-20x em memória. Use ao treinar/executar transformers com sequências longas (>512 tokens), ao encontrar problemas de memória GPU com atenção, ou quando precisa de inferência mais rápida. Suporta SDPA nativo do PyTorch, biblioteca flash-attn, H100 FP8 e sliding window attention.
日本語の概要は準備中です。原文の説明を表示しています。
Post-training 4-bit quantization for LLMs with minimal accuracy loss. Use for deploying large models (70B, 405B) on consumer GPUs, when you need 4× memory reduction with <2% perplexity degradation, or for faster inference (3-4× speedup) vs FP16. Integrates with transformers and PEFT for QLoRA fine-tuning.
日本語の概要は準備中です。原文の説明を表示しています。
Activation-aware weight quantization for 4-bit LLM compression with 3x speedup and minimal accuracy loss. Use when deploying large models (7B-70B) on limited GPU memory, when you need faster inference than GPTQ with better accuracy preservation, or for instruction-tuned and multimodal models. MLSys 2024 Best Paper Award winner.
日本語の概要は準備中です。原文の説明を表示しています。
Systematic performance profiling and optimization methodology. Covers CPU profiling, memory profiling, heap analysis, goroutine/thread analysis, FlameGraph generation, latency profiling, database query profiling, network I/O profiling, and continuous performance regression detection. USE WHEN: investigating performance issues, optimizing slow code, reducing memory usage, diagnosing GC/CPU bottlenecks, finding N+1 queries, analyzing pprof/flamegraph output, or establishing a performance baseline. Triggers on "profiling", "performance analysis", "slow code", "flame graph", "pprof", "heap dump", "memory leak", "CPU spike", "bottleneck".
日本語の概要は準備中です。原文の説明を表示しています。
Spark, pandas, polars, DuckDB optimization for batch data processing. Activate on: batch processing, Spark optimization, polars, DuckDB, pandas performance, data frame, shuffle, partition, memory optimization. NOT for: streaming pipelines (use streaming-pipeline-architect), warehouse queries (use data-warehouse-optimizer).
日本語の概要は準備中です。原文の説明を表示しています。
Optimize AgentDB performance with quantization (4-32x memory reduction), HNSW indexing (150x faster search), caching, and batch operations. Use when optimizing memory usage, improving search speed, or scaling to millions of vectors.
日本語の概要は準備中です。原文の説明を表示しています。
Achieve aggressive v3 performance targets: 2.49x-7.47x Flash Attention speedup, 150x-12,500x search improvements, 50-75% memory reduction. Comprehensive benchmarking and optimization suite.
日本語の概要は準備中です。原文の説明を表示しています。
Achieve aggressive v3 performance targets: 2.49x-7.47x Flash Attention speedup, 150x-12,500x search improvements, 50-75% memory reduction. Comprehensive benchmarking and optimization suite.
日本語の概要は準備中です。原文の説明を表示しています。
Optimize AgentDB performance with quantization (4-32x memory reduction), HNSW indexing (150x faster search), caching, and batch operations. Use when optimizing memory usage, improving search speed, or scaling to millions of vectors.
日本語の概要は準備中です。原文の説明を表示しています。
Optimize Apache Spark jobs with partitioning, caching, shuffle optimization, and memory tuning. Use when improving Spark performance, debugging slow jobs, or scaling data processing pipelines.
日本語の概要は準備中です。原文の説明を表示しています。
Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.
日本語の概要は準備中です。原文の説明を表示しています。