sglang-diffusion-add-model

Generate SGLang integration templates for Diffusers-based diffusion models.

Updated May 14, 2026
One-click install
npx skills add https://github.com/sqjian/sglang --skill sglang-diffusion-add-model-sqjian
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sglang-diffusion-add-model
Source: https://github.com/sqjian/sglang/tree/main/python/sglang/multimodal_gen/.claude/skills/sglang-diffusion-add-model
Command: npx skills add https://github.com/sqjian/sglang --skill sglang-diffusion-add-model-sqjian

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates the onboarding of a new diffusion model or Diffusers pipeline into SGLang by providing a structured scaffold of config, pipelines, and stages, reducing boilerplate and ensuring consistency across projects.

Core Features & Use Cases

  • Hybrid/Modular templates for new diffusion models to speed up integration with the SGLang runtime.
  • Scaffold files: creates DiT/vAE configs, pipeline skeletons, registry entries, and BeforeDenoisingStage to standardize onboarding.
  • Use Case: A researcher adds a new Diffusers-based model; the skill generates the necessary templates and hooks to run within SGLang with minimal manual edits.

Quick Start

Provide a diffusion model repository path or Diffusers pipeline reference; the skill will scaffold integration templates and configuration for SGLang.

Frequently Asked Questions about sglang-diffusion-add-model

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

FAQPage Schema
How do I integrate a Diffusers pipeline into SGLang?

Scaffold a new diffusion model integration into SGLang by generating modular templates for DiT configs, pipeline skeletons, and registry entries, reducing boilerplate when onboarding Diffusers-based models.

What do I need to provide to scaffold a diffusion model in SGLang?

You must supply a Diffusers-compatible model repository path or pipeline reference to generate the correct config templates, pipeline skeletons, and registry wiring for SGLang.

Does the SGLang diffusion integration skill generate registry entries automatically?

Yes, scaffolding a diffusion model in SGLang automatically generates registry entries alongside DiT/vAE configs, pipeline skeletons, and BeforeDenoisingStage hooks to ensure consistent onboarding.

Can I use this scaffolding for any Diffusers-based model?

Yes, you can scaffold any Diffusers-based model integration by providing its repository reference, which generates the necessary templates, configs, and hooks to run within SGLang.

What is the best way to standardize diffusion model onboarding in SGLang?

Use automated scaffolding to standardize diffusion model onboarding in SGLang, generating structured configs, pipeline stages, and registry entries to ensure repeatable integration workflows across projects.