Loading...
Loading...
Found 33 Skills
This skill provides comprehensive guidance for adapting Wan-series video generation models (Wan2.1/Wan2.2) from NVIDIA CUDA to Huawei Ascend NPU. It should be used when performing NPU migration of DiT-based video diffusion models, including device layer adaptation, operator replacement, distributed parallelism refactoring, attention optimization, VAE parallelization, and model quantization. This skill covers 9 major adaptation domains derived from real-world Wan2.2 CUDA-to-Ascend porting experience.
Apply quantization to reduce memory by 4-32x. Enable HNSW indexing for 150x faster search. Configure caching strategies and implement batch operations. Use when optimizing memory usage, improving search speed, or scaling to millions of vectors. Deploy these optimizations to achieve 12,500x performance gains.
Convert HuggingFace transformer models to ONNX format for browser inference with Transformers.js and WebGPU. Use when given a HuggingFace model link to convert to ONNX, when setting up optimum-cli for ONNX export, when quantizing models (fp16, q8, q4) for web deployment, when configuring Transformers.js with WebGPU acceleration, or when troubleshooting ONNX conversion errors. Triggers on mentions of ONNX conversion, Transformers.js, WebGPU inference, optimum export, model quantization for browser, or running ML models in the browser.
Memory-efficient fine-tuning with 4-bit quantization and LoRA adapters. Use when fine-tuning large models (7B+) on consumer GPUs, when VRAM is limited, or when standard LoRA still exceeds memory. Builds on the lora skill.
Runs LLM inference on CPU, Apple Silicon, and consumer GPUs without NVIDIA hardware. Use for edge deployment, M1/M2/M3 Macs, AMD/Intel GPUs, or when CUDA is unavailable. Supports GGUF quantization (1.5-8 bit) for reduced memory and 4-10× speedup vs PyTorch on CPU.
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.
vLLM Ascend plugin for LLM inference serving on Huawei Ascend NPU. Use for offline batch inference, API server deployment, quantization inference (with msmodelslim quantized models), tensor/pipeline parallelism for distributed serving, and OpenAI-compatible API endpoints. Supports Qwen, DeepSeek, GLM, LLaMA models with Ascend-optimized kernels.
Create and manage Neo4j vector indexes, run vector similarity search (ANN/kNN), store embeddings on nodes or relationships, use SEARCH clause (Neo4j 2026.01+, preferred) or db.index.vector.queryNodes() procedure (deprecated 2026.04, still works on 2025.x), configure HNSW and quantization options, pick similarity function and embedding provider dimensions, and batch-update embeddings. Use when tasks involve CREATE VECTOR INDEX, vector.dimensions, cosine/euclidean search, embedding ingestion pipelines, or semantic nearest-neighbor lookup. Does NOT handle GraphRAG retrieval_query graph traversal — use neo4j-graphrag-skill. Does NOT handle fulltext/keyword indexes (FULLTEXT INDEX, db.index.fulltext) — use neo4j-cypher-skill. Does NOT handle GDS graph embeddings (FastRP, Node2Vec) — use neo4j-gds-skill.
PyTorch-based TAO image classification. Supports a wide range of backbones (FAN, EfficientNet, ResNet, etc.) with distillation and quantization for deployment. Use when training, evaluating, distilling, quantizing, exporting, or running inference for a TAO image-classification (PyT) model. Trigger phrases include "train image classifier", "TAO classification", "ResNet/EfficientNet/FAN backbone classifier", "classification-pyt".
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.
PyTorch implementation of TurboQuant for LLM KV cache compression using two-stage vector quantization (random rotation + Lloyd-Max + QJL residual correction).
GGUF format and llama.cpp quantization for efficient CPU/GPU inference. Use when deploying models on consumer hardware, Apple Silicon, or when needing flexible quantization from 2-8 bit without GPU requirements.