npu-op-benchmark
CommunityStandardized NPU operator performance benchmarking tool.
Authordongg622
Version1.0.0
Installs0
System Documentation
What problem does it solve?
This Skill provides a reliable method to measure the performance of individual operators in Ascend NPU environments, aiding in performance optimization and hardware assessment.
Core Features & Use Cases
- Environment Compatibility Check: Verifies CANN, torch, and torch_npu versions and validates supported configurations.
- Performance Testing: Executes burst testing on specified operators like repeat_interleave across different hardware setups such as Docker containers or local conda environments.
- Use Case: Use this Skill to benchmark the latency of a specific operator in your AI inference pipeline to identify bottlenecks or compare hardware capabilities.
Quick Start
Provide server details, environment setup, and target operator to run an automated performance benchmark.
Dependency Matrix
Required Modules
None requiredComponents
scriptsreferences
💻 Claude Code Installation
Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.
Please help me install this Skill: Name: npu-op-benchmark Download link: https://github.com/dongg622/china-ai-chip-skill/archive/main.zip#npu-op-benchmark Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
Agent Skills Search Helper
Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.