skills.rest
Nabilhassan12345
Nabilhassan12345COMMUNITY
Explore all skills by this author →
GitHub

sglang-diffusion-benchmark-profile

Benchmark and profile SGLang Diffusion denoising pipelines for latency and resource usage.

Updated Apr 7, 2026
Search Tags:#profiling#latency#benchmark#gpu#diffusion#sglang#torch.profiler
One-click install
npx skills add https://github.com/Nabilhassan12345/voice-ai-workspace --skill sglang-diffusion-benchmark-profile
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: sglang-diffusion-benchmark-profile
Source: https://github.com/Nabilhassan12345/voice-ai-workspace/tree/main/sglang/python/sglang/multimodal_gen/.claude/skills/sglang-diffusion-benchmark-profile
Command: npx skills add https://github.com/Nabilhassan12345/voice-ai-workspace --skill sglang-diffusion-benchmark-profile

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Diffusion model benchmarks and profiling streamline the process of measuring denoise latency, end-to-end generation timing, and memory usage, enabling data-driven optimization decisions.

Core Features & Use Cases

  • End-to-end diffusion benchmarks: measure denoise latency, end-to-end latency, and peak memory across presets.
  • Profiling workflow: collect torch.profiler traces and perf dumps to rank hot kernels.
  • Guided optimization handoff: map hotspots to existing fast paths and hand off kernel work to specialized optimization skills when needed.

Quick Start

Run the diffusion benchmark profile for a chosen model to measure denoise latency and collect profiling data.

Frequently Asked Questions about sglang-diffusion-benchmark-profile

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I benchmark diffusion model denoising latency on a GPU?▼

You can profile diffusion latency by collecting torch.profiler traces and perf dumps to rank hot kernels and quantify resource usage during denoising pipelines.

How does profiling diffusion latency help identify bottlenecks in SGLang workflows?▼

Profiling diffusion latency identifies bottlenecks by mapping hotspots from torch.profiler traces to existing fast paths, enabling guided optimization handoffs to specialized kernel work.

What is needed to set up a reproducible environment for diffusion benchmarking?▼

Setting up reproducible diffusion benchmarking requires completing preflight steps to ensure accurate profiling data collection and consistent denoise latency measurement.

Can I collect torch.profiler traces for diffusion models across different presets?▼

Yes, you can collect torch.profiler traces across multiple presets and models to support performance comparisons and rank hot kernels within diffusion workflows.

What is the best way to analyze peak memory usage during diffusion generation?▼

The best way to analyze peak memory is running end-to-end diffusion benchmarks that measure memory across presets to support data-driven optimization decisions.

Related Skills

Explore More Software Engineering Skills→
sqjiansqjian

sglang-diffusion-performance

Benchmark and tune SGLang diffusion performance configurations across GPUs and VRAM budgets.

Community
Advanced
hey-konghey-kong

sglang-diffusion-benchmark-profile

Benchmark denoise latency and profile hotspots in SGLang diffusion backends.

Community
Advanced
sgl-projectsgl-project
31.6k

diffusion-kernel

Integrate custom JIT CUDA and Triton kernels to accelerate diffusion model inference.

Official
Advanced
QingHuan-6QingHuan-6

sglang-diffusion-benchmark-profile

Measure diffusion denoise latency and collect performance traces for SGLang.

Community
Advanced